System

Abstract

To provide a system.SOLUTION: A system includes: first means for learning a record of a specific person, which is data including information on intelligence, thinking, feeling, emotion, morals, and a sense of ethics with respect to the specific person; second means for reproducing a thinking clone of the specific person on the basis of the learned record of the specific person; third means for acquiring a question of a user with respect to the thinking clone of the specific person; fourth means for generating a first prompt sentence for outputting an answer from the thinking clone of the specific person to the question on the basis of contents of the question by the user; fifth means for inputting the generated first prompt sentence into a generative AI model to generate an answer from the thinking clone of the specific person; and sixth means for transmitting the generated answer from the thinking clone of the specific person to a terminal of the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] With current technology, it is difficult to understand a specific person's thoughts and principles of behavior and predict how that person will react to future events. [Means for solving the problem]

[0005] This invention provides a means to study the records of a specific person and recreate a thought clone of that person. By asking questions to the recreated personality, it is possible to understand and verify the person's past behavioral principles. Furthermore, by providing the personality with current information, it is possible to predict how the person will speak / act in response to future events. [Brief explanation of the drawings]

[0006] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0008] First, the terms used in the following description will be explained.

[0009] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be one arithmetic device or a combination of multiple arithmetic devices. Also, the processor may be one type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit),

[0010] Examples include a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0011] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0012] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g.,

[0013] hard disk), or magnetic tape.

[0014] In the following embodiments, the coded communication I / F (Interface) refers to a communication processor and

[0015] The communication I / F is an interface that includes a network interface (NIC) and an antenna. The communication I / F controls communication between multiple computers. Examples of communication standards that can be applied to the communication I / F include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), and Bluetooth (registered trademark).

[0016] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0017] [First embodiment]

[0018] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0019] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0020] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0021] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0022] The reception device 38 is equipped with a touch panel 38A and a microphone 38B.

[0023] The touch panel 38A receives user input by touch of a pointer (for example, a pen or a finger) by detecting contact with the pointer. The microphone 38B receives user input by voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0024] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 includes an optical system including a lens, an aperture, and a shutter, and a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or CC

[0025] A compact digital camera equipped with an imaging element such as a D (Charge Coupled Device) image sensor.

[0026] It's a camera.

[0027] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0028] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0032] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0033] "Example 1"

[0034] In one embodiment of the present invention, a machine learning model is developed to learn the records of a specific individual by collecting digital data such as the individual's past actions, statements, books, papers, and social media posts. This data is then input into a machine learning model that learns the individual's thought patterns and principles of behavior, enabling the reproduction of that individual's personality.

[0035] "Example 2"

[0036] As a means of asking questions to the recreated personality, natural language processing technology is used to provide an interface that allows humans to ask questions naturally. For example, by asking a question such as, "What do you think about smartphones, which didn't exist in your lifetime?", it is possible to predict how the person will react to modern technology.

[0037] "Example 3"

[0038] To provide current information to the personality, information such as the latest news articles and research papers is input into the system, which can then predict how the person will react to current social conditions and advances in science and technology.

[0039] The processing flow of each embodiment will be described below.

[0040] "Example 1"

[0041] Step 1: Collect digital data such as a specific person's past actions, statements, works such as books and papers, and social media posts.

[0042] Step 2: Preprocess the collected data and convert it into a format suitable for the machine learning model.

[0043] Step 3: The preprocessed data is input into the machine learning model and the model is trained. As a result of the training, a model that has learned a person's thought patterns and principles of behavior is obtained.

[0044] "Example 2"

[0045] Step 1: Using natural language processing technology, develop an interface that allows humans to ask questions naturally.

[0046] Step 2: Ask the recreated personality questions through the developed interface. Step 3: Analyze the responses from the recreated personality and predict how the person will react to modern technology.

[0047] "Example 3"

[0048] Step 1: Gather information such as the latest news articles and research papers.

[0049] Step 2: Enter the collected information into the system.

[0050] Step 3: Based on the input information, predict how the recreated personality will react to current social conditions and advances in science and technology.

[0051] Example 1

[0052] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0053] Conventional systems have had difficulty accurately reproducing a specific person's thought patterns and principles of behavior. Furthermore, preprocessing of collected data and building, training, and evaluating machine learning models were often done manually, resulting in inefficiencies. Furthermore, the process of posing appropriate questions to the reproduced personality and generating responses was complicated.

[0054] The identification process by the identification processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means. In this invention, the server includes a means for collecting digital data and performing preprocessing, a means for constructing and training a machine learning model, a means for evaluating the trained model, and a means for inputting a prompt sentence into the generative AI model and generating a response. This makes it possible to efficiently and accurately reproduce the thought patterns and principles of behavior of a specific person, ask appropriate questions to the reproduced personality, and generate a response.

[0055] A "specific person" refers to a specific individual, and is the subject of collection of digital data such as that individual's past actions, statements, copyrighted works, and social media posts.

[0056] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0057] "Thought clone" refers to a machine learning model built to replicate a specific person's thought patterns and behavioral principles.

[0058] "Digital data" refers to electronically stored information such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0059] "Preprocessing" refers to the process of cleaning, tokenizing, normalizing, and otherwise converting collected digital data into a format suitable for machine learning models.

[0060] A "machine learning model" refers to an algorithm or neural network used to learn and replicate a specific person's thought patterns and behavioral principles.

[0061] "Learning" refers to the process of inputting data into a machine learning model so that the model can understand and reproduce a specific person's thought patterns and principles of behavior.

[0062] "Evaluation" refers to the testing and verification process to confirm the predictive accuracy and performance of a trained machine learning model.

[0063] "Generative AI model" refers to an artificial intelligence model that is used to reproduce a specific person's thought patterns and principles of behavior and generate responses to prompts.

[0064] A "prompt" refers to a question or instruction entered into a generative AI model, and is input data used to generate a response based on a specific person's thought patterns.

[0065] This invention is a system that learns the records of a specific person and reproduces that person's thought patterns and principles of behavior. A specific embodiment of this system will be described below.

[0066] First, the user collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. This data serves as the basis for learning the person's thought patterns and principles of behavior. The collected data is then sent to a server.

[0067] The server then preprocesses the collected digital data, specifically cleaning, tokenizing, and normalizing the text data. The software used includes Python libraries. For example, this includes removing unnecessary HTML tags from social media posts and segmenting the text into words and sentences.

[0068] Using the preprocessed data, the server builds a machine learning model. This model is used to learn the thought patterns and behavioral principles of a specific person. Specifically, a deep learning framework such as TENSORFLOW (registered trademark) is used. Model construction includes dividing the dataset, selecting a model, and designing a neural network.

[0069] The server then inputs the preprocessed data into the machine learning model it has built and trains the model. The model is trained using a training set and an optimization algorithm such as gradient descent is used to minimize the loss function.

[0070] Once the training is complete, the server evaluates the model. The evaluation checks the model's predictive accuracy using a test set. Evaluation metrics include precision, recall, and F1 score.

[0071] Finally, the user uses the trained model to recreate a specific person's personality. Specifically, the user inputs a prompt into the generative AI model, which simulates how the specific person would respond. For example, the user can input the following prompt:

[0072] "Tell me about your latest book."

[0073] The generative AI model recreates the thought patterns of a specific person based on the input prompt and generates a response that is highly realistic because it is based on the specific person's past behavioral principles and thought patterns.

[0074] The above is a specific embodiment for carrying out the present invention. This system makes it possible to efficiently and accurately reproduce the thought patterns and behavioral principles of a specific person, pose appropriate questions to the reproduced personality, and generate responses.

[0075] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0076] Step 1: Data collection

[0077] A user collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. The input is digital data collected from the internet or databases. The output is the collected digital data sent to a server. Specifically, a web scraping tool is used to automatically collect social media posts.

[0078] Step 2: Data Preprocessing

[0079] The server preprocesses the collected digital data. The input is the collected digital data. The output is the preprocessed data. Specific operations include cleaning, tokenizing, and normalizing the text data. For example, a Python library is used to tokenize the text and remove unnecessary HTML tags.

[0080] Step 3: Building a machine learning model

[0081] The server builds a machine learning model using the preprocessed data. The input is the preprocessed data. The output is the built machine learning model. Specifically, it designs a neural network model using a deep learning framework such as TensorFlow. It splits the dataset into a training set and a test set, and selects and designs a model.

[0082] Step 4: Training the model

[0083] The server inputs the preprocessed data into the constructed machine learning model and trains the model. The inputs are the preprocessed data and the constructed model. The output is a trained model. Specifically, the model is trained using a training set, and an optimization algorithm such as gradient descent is used to minimize the loss function.

[0084] Step 5: Evaluate the model

[0085] The server evaluates the model after training. The inputs are the trained model and a test set. The output is the model's predictive accuracy and performance. Specifically, it checks the model's predictive accuracy using the test set and calculates evaluation metrics such as precision, recall, and F1 score.

[0086] Step 6: Personality Regeneration

[0087] The user uses a trained model to recreate the personality of a specific person. The inputs are the trained model and a prompt. The output is a generated response. Specifically, the prompt is input into the generative AI model, and a simulation is performed to see how the specific person would respond. For example, the prompt is input as "Tell me about your latest book," and the model generates a response.

[0088] (Application example 1)

[0089] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0090] In today's world, there is a demand for systems that can reproduce the knowledge and thought patterns of specific famous people or historical figures, allowing users to converse with them. However, with conventional technology, it is difficult to effectively collect digital data about a specific person, input it into a machine learning model, and generate answers to user questions that are appropriate for that person. Therefore, the challenge is to provide a system that can accurately reproduce the thought patterns and principles of behavior of a specific person, allowing users to converse with that person.

[0091] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0092] In this invention, the server includes a means for learning the records of a specific person, a means for reproducing a thought clone of the person based on the learned records, and a means for asking questions to the reproduced personality. This makes it possible to collect digital data of a specific person, input it into a machine learning model for learning, and generate answers to user questions that are characteristic of that person.

[0093] "Specific person" refers to a specific individual, such as a particular celebrity or historical figure.

[0094] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0095] A "thinking clone" refers to an artificial intelligence model that reproduces the thought patterns and behavioral principles of a specific person.

[0096] A "machine learning model" refers to an algorithm or system that learns a specific person's thought patterns and behavioral principles based on collected digital data.

[0097] "Digital data" refers to electronically stored information such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0098] "User" means an individual who uses the System to interact with a specific person.

[0099] "Questions" refer to questions that the user poses to the recreated personality.

[0100] "Answer" refers to the response that the recreated personality generates to the user's question.

[0101] A "thinking pattern" refers to how a particular person tends to think and judge a particular situation or problem.

[0102] A "principle of action" refers to a basic principle of how a particular person should act in a particular situation or problem.

[0103] To implement this invention, it is necessary to build a system that collects digital data of specific individuals and inputs it into a machine learning model for learning. This system consists of three main elements: a server, a terminal, and a user.

[0104] server

[0105] The server includes a means for learning the records of a specific person, a means for reproducing a thought clone of that person based on the learned records, and a means for asking questions to the reproduced personality. Specifically, the server performs the following processes.

[0106] 1. Data collection: The server collects digital data such as social media posts, copyrighted works, and statements of specific individuals. This data is collected automatically using APIs.

[0107] 2. Data input: Convert the collected data into JSON format and input it into a machine learning model such as OpenAI's GPT-3.

[0108] 3. Model learning: The GPT-3 model learns a specific person's thought patterns and behavioral principles based on the input data.

[0109] 4. Response generation: The trained model is used to generate a response that is appropriate for the user's question.

[0110] Terminal

[0111] A terminal provides an interface for users to interact with the system. Devices such as smartphones and tablets fall into this category. Terminals have the following functions:

[0112] 1. User Interface: Provides an interface for users to enter questions and receive answers.

[0113] 2. Data transmission: The question entered by the user is sent to the server.

[0114] 3. Data reception: Receives the response sent from the server and displays it to the user.

[0115] User

[0116] A user is an individual who uses the system to interact with a specific person. The user performs the following operations:

[0117] 1. Question input: The user inputs a question for a specific person through the terminal interface.

[0118] 2. Receiving the answer: The user receives the answer sent from the server on their terminal and checks it.

[0119] Specific examples

[0120] For example, if a user selects "famous scientists" as a specific person, the user can enter a question such as:

[0121] "Tell me about your next project."

[0122] "Tell me about the theory of relativity."

[0123] The server uses the trained model to generate answers to these questions and provides them to the user via the terminal.

[0124] Prompt Sentence Examples

[0125] "I discovered something new today."

[0126] I'm thinking about my next project.

[0127] "Science is constantly evolving."

[0128] In this way, the user can experience the knowledge and thought patterns of a particular person and enjoy the dialogue.

[0129] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0130] Step 1:

[0131] The server collects digital data of a specific person. Specifically, it automatically collects data such as social media posts, copyrighted works, and statements using an API. The input is the person's identifying information, and the output is the collected digital data.

[0132] Step 2:

[0133] The server converts the collected digital data into JSON format. The input is the collected digital data, and the output is JSON format data. This conversion makes the data suitable for machine learning models.

[0134] Step 3:

[0135] The server inputs the JSON-formatted digital data into a machine learning model such as OpenAI's GPT-3, training the model. The input is JSON-formatted data, and the output is a trained model. The model learns a specific person's thought patterns and principles of behavior.

[0136] Step 4:

[0137] A user inputs a question for a specific person through the terminal interface. The input is the user's question, and the output is the question data sent from the terminal to the server.

[0138] Step 5:

[0139] The terminal sends the question entered by the user to the server. The input is the user's question data, and the output is the data to be sent to the server.

[0140] Step 6:

[0141] The server generates answers to user questions using the trained model. The input is the user's question data and the trained model, and the output is the generated answer. The server inputs the question as a prompt to the model and generates an answer.

[0142] Step 7:

[0143] The server sends the generated answer to the terminal. The input is the generated answer, and the output is the data sent to the terminal.

[0144] Step 8:

[0145] The terminal receives the response sent from the server and displays it to the user. The input is the data sent from the server, and the output is the response displayed to the user.

[0146] In this way, the user can experience the knowledge and thought patterns of a particular person and enjoy the dialogue.

[0147] Example 2

[0148] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0149] In conventional systems, the process of asking questions to the reproduced personality and receiving responses was complicated, making it difficult for users to ask questions naturally. Also, it was difficult to predict how the reproduced personality would respond to modern technology and information.

[0150] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a question, a means for a terminal to send a question to the server, a means for the server to analyze the question, a means for the server to generate a prompt sentence for the generative AI model, a means for the server to input a prompt sentence to the generative AI model, a means for the server to receive a generated response, a means for the server to send the response to the terminal, and a means for the terminal to display the response to the user. This allows the user to naturally ask a question and easily obtain a response from the reproduced personality to the question.

[0151] "Records of a specific person" are data that include information about a specific person's intelligence, thoughts, feelings, emotions, morals, ethics, etc.

[0152] "Means of learning" are methods and techniques for analyzing the records of a particular person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[0153] "Means for reproducing thought clones" refers to methods and techniques for generating artificial intelligence that imitates the thoughts and actions of a specific person based on the records of that person that it has learned.

[0154] A "means for asking a question" is a method or technique that allows a user to naturally input a question to the reproduced personality and transmit the question to the system.

[0155] "Means for understanding and verifying the principles of past behavior" refers to methods and techniques for analyzing the past behavior and thought patterns of the recreated personality and predicting responses to current questions based on that analysis.

[0156] "Current means of providing information" are methods and techniques for providing modern technology and information to the recreated personality and eliciting its response.

[0157] "Means of eliciting how to speak / act in response to possible future events" refers to methods or techniques for predicting how the recreated personality will react to future events and generating that response.

[0158] The "means for the user to input a question" is an interface that allows the user to input a question to the reproduced personality using a terminal.

[0159] The "means by which the terminal transmits the question to the server" refers to the communication protocol or technology used to transmit the question entered by the user from the terminal to the server.

[0160] The "means by which the server analyzes the question" refers to a method or technology by which the server analyzes the question received by using natural language processing technology and extracts the grammatical structure and keywords of the question.

[0161] "Means for generating prompt sentences for a generative AI model" refers to methods or technologies for generating appropriate prompt sentences for a generative AI model based on the analysis results.

[0162] "Means for inputting prompt sentences into a generative AI model" refers to methods or techniques for inputting generated prompt sentences into a generative AI model and obtaining a response from the model.

[0163] "Means for receiving the generated response" refers to the method or technology by which the server receives the response from the generative AI model.

[0164] The "means for sending the response to the terminal" refers to the communications protocol or technology by which the server sends the generated response to the user's terminal.

[0165] The "means by which the terminal displays the response to the user" is an interface for displaying the response received by the terminal to the user.

[0166] This invention is a system for posing questions to a recreated personality and obtaining responses. The system includes means for learning a record of a specific person and recreating a thought clone based on the record. The system also includes means for posing questions to the recreated personality and generating responses to the questions.

[0167] Hardware and software used

[0168] Hardware: Servers, user devices (PCs, smartphones, etc.)

[0169] Software: Natural language processing libraries, generative AI models (e.g., GPT-4 (registered trademark))

[0170] Specific operation of the system

[0171] 1. User enters a question:

[0172] The user uses the device to input questions for the recreated personality, such as "What do you think about smartphones, which didn't exist in your lifetime?"

[0173] 2. The device sends a question to the server:

[0174] The terminal sends the question entered by the user to the server using a communication protocol such as an HTTP request.

[0175] 3. The server parses the question:

[0176] The server analyzes the received question using a natural language processing library. Specifically, it analyzes the grammatical structure of the question and extracts keywords. For example, keywords such as "smartphone" and "What do you think?" are extracted.

[0177] 4. The server generates a prompt for the generative AI model:

[0178] Based on the analysis results, the server generates a prompt for the generative AI model (e.g., GPT-4). For example, it generates a prompt such as, "You are a person from the 19th century. What do you think about modern smartphones?"

[0179] 5. The server inputs a prompt to the generative AI model:

[0180] The server inputs the generated prompt sentence into the generative AI model, which then generates a response based on the prompt sentence.

[0181] 6. The server receives the generated response:

[0182] The server receives a response from the generative AI model, such as, "I think smartphones are very useful tools, but they were unimaginable in my day."

[0183] 7. The server sends the response to the device:

[0184] The server then sends the generated response to the user's device, again using a communication protocol such as an HTTP response.

[0185] 8. The terminal displays the response to the user:

[0186] The terminal displays the received response to the user, who can then check the response from the personality played back on the screen.

[0187] Specific examples

[0188] User question: "What do you think about smartphones, which didn't exist in your lifetime?"

[0189] Prompt for generative AI model: "You are a person from the 19th century. What do you think about modern smartphones?"

[0190] This system allows users to naturally ask questions to the reproduced personality and easily obtain responses to those questions.

[0191] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0192] Step 1:

[0193] The user enters a question.

[0194] The user inputs a question using the device's keyboard or touch screen. Once input is complete, the user clicks the "Submit" button. The input data is a text-based question. For example, a question might be entered such as, "What do you think about smartphones, which didn't exist when you were alive?"

[0195] Step 2:

[0196] The terminal sends a question to the server.

[0197] The device converts the input question into JSON format and sends it to the server as an HTTP POST request. The input data is the question in text format, and the output data is the question data in JSON format.

[0198] Step 3:

[0199] The server parses the question.

[0200] The server parses the received JSON data and extracts the question. It uses a natural language processing library to analyze the grammatical structure of the question and extract keywords. The input data is the question data in JSON format, and the output data is the analysis result (grammatical structure and keywords). For example, keywords such as "smartphone" and "What do you think?" are extracted.

[0201] Step 4:

[0202] The server generates a prompt for the generative AI model.

[0203] Based on the analysis results, the server generates a prompt for the generative AI model (e.g., GPT-4). The input data is the analysis results, and the output data is the prompt. For example, it generates a prompt such as, "You are a person from the 19th century. What do you think about modern smartphones?"

[0204] Step 5:

[0205] The server inputs a prompt sentence into the generative AI model.

[0206] The server inputs the generated prompt sentence into the generative AI model and waits for the model's response. The input data is the prompt sentence, and the output data is the response from the generative AI model.

[0207] Step 6:

[0208] The server receives the generated response.

[0209] The server receives the response from the generative AI model and checks the response. The input data is the response from the generative AI model, and the output data is the checked response. For example, a response such as "I think smartphones are very useful tools, but they were unimaginable in my day" is generated.

[0210] Step 7:

[0211] The server sends a response to the terminal.

[0212] The server converts the generated response back into JSON format and sends it to the user's device as an HTTP response. The input data is the confirmed response, and the output data is the JSON-formatted response data.

[0213] Step 8:

[0214] The terminal displays the response to the user.

[0215] The device parses the received JSON data and displays the response on the screen. The input data is the response data in JSON format, and the output data is the response in text format. The user can check the displayed response and enjoy interacting with the reproduced personality.

[0216] (Application example 2)

[0217] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0218] Previous systems had issues with the interactive experience, as it was difficult for users to ask questions naturally to the reproduced personality and there was no function to play back the answers aloud. Furthermore, the user experience was limited because there was no way for users to ask questions by voice or text input.

[0219] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of the person based on the learned records, means for posing questions to the reproduced personality, means for understanding and verifying the personality's past behavioral principles, means for providing the personality with current information, means for eliciting how the personality will speak / act in response to potential future events, means for reproducing the generated answers aloud using speech synthesis technology, and means for the user to pose questions by voice input or text input. This allows the user to naturally pose questions and reproduce the generated answers aloud, enabling a more interactive and richer experience.

[0220] "Records of a specific person" refers to information about a specific person, including their intelligence, thoughts, feelings, emotions, morals, and ethics.

[0221] "Means of learning" refers to techniques for analyzing records of a specific person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[0222] A "thought clone" refers to an artificial intelligence that imitates the intelligence, thoughts, emotions, feelings, morals and ethics of a specific person based on learned records.

[0223] "Means for asking questions" refers to an interface that allows the user to naturally ask questions to the recreated personality.

[0224] "Past behavioral principles" refers to principles or patterns that show how a particular person has thought or acted in the past.

[0225] "Current information" refers to data about modern technology and events.

[0226] "Means of eliciting what to say / do" refers to technology that predicts how the recreated personality will react to future events and provides that reaction to the user.

[0227] "Speech synthesis technology" refers to the technology that converts text data into speech.

[0228] "Means for playing back by voice" refers to technology for providing the generated answer to the user by voice using speech synthesis technology.

[0229] "Voice or text input means for asking questions" refers to an interface that allows a user to ask questions to the played-out persona using voice or text.

[0230] The system for implementing this invention includes a program for learning the records of a specific person and recreating a thought clone based on the records. The system allows the user to naturally ask questions to the regenerated personality and provides a function for replaying the answers by voice.

[0231] Hardware and software used

[0232] Hardware: Smartphone

[0233] Software: Python, OpenAI API (GPT-3), speech synthesis library

[0234] Data processing and calculation

[0235] 1. User input: The user uses their smartphone to ask a question by voice or text input.

[0236] 2. Question processing: The question is sent to the server, and an answer is generated for the specified personality using OpenAI's GPT-3 model.

[0237] 3. Answer Generation: The GPT-3 model generates appropriate answers to questions based on records of a specific person.

[0238] 4. Speech synthesis: The generated answer is converted into speech using a library.

[0239] 5. Audio playback: The answer is played aloud to the user through the smartphone speaker.

[0240] Specific examples

[0241] A user opens the app and asks a question to a specific historical figure. For example, "What do you think about smartphones, which didn't exist in your time?" This question is sent to a server, and the GPT-3 model generates an answer for that person. The answer might be something like, "Smartphones allow instant access to information and will greatly contribute to scientific advancement." This answer is then played aloud using speech synthesis technology and provided to the user.

[0242] Prompt Sentence Examples

[0243] As a specific historical figure, answer the following question: What do you think about smartphones, which didn't exist in your lifetime?

[0244] In this way, users can naturally ask questions and have the generated answers played back to them aloud, resulting in a more interactive and enriching experience.

[0245] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0246] Step 1:

[0247] A user starts the application using a smartphone and selects a specific historical figure. Next, the user inputs a question by voice or text. This input data is text data containing the user's question.

[0248] Step 2:

[0249] The device sends the question entered by the user to the server. The server analyzes the received question and sends a prompt sentence to the generative AI model (GPT-3) to generate an answer as a specific historical figure. This prompt sentence is text data containing the user's question.

[0250] Step 3:

[0251] The server receives a response from the generative AI model (GPT-3), which generates an answer to the question based on the records of specific historical figures. This generated answer is returned to the server as text data.

[0252] Step 4:

[0253] The server uses a speech synthesis library (pyttsx3) to convert the received text data into speech data. The speech synthesis library analyzes the text data and generates the corresponding speech data.

[0254] Step 5:

[0255] The server transmits the generated voice data to the device, which then plays the received voice data and provides it to the user. The user can then listen to the audio of the specific historical figure's response through the smartphone speaker.

[0256] Step 6:

[0257] The user can then enter the question again or select a different historical figure, allowing the user to continue the interactive experience.

[0258] Example 3

[0259] Next, a third embodiment of the third embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0260] While conventional systems can reproduce a specific person's past actions and thoughts, it is difficult to predict how that person will react to the latest information. Furthermore, there is a lack of concrete means for inputting information such as the latest news articles or research papers and predicting how a specific person will react to that information. This makes it difficult to accurately predict a specific person's future actions and statements.

[0261] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0262] In this invention, the server includes means for learning the records of a specific person, means for recreating a thought clone of the person based on the learned records, means for asking questions to the recreated personality, means for understanding and verifying the personality's past behavioral principles, means for providing current information to the personality, means for analyzing the information, means for generating prompt sentences based on the analysis results, means for inputting the generated prompt sentences into a generative AI model, means for obtaining prediction results from the generative AI model, and means for deriving how the personality will speak / act in response to possible future events. This makes it possible to predict how a specific person will react to the latest information.

[0263] A "specific person" is an individual person that the system targets, and is the subject of learning records of that person's past actions, thoughts, emotions, etc.

[0264] "Means of studying records" refers to techniques and methods for collecting information, including a specific person's intelligence, thoughts, feelings, emotions, morals, and ethics, and for analyzing and studying this information.

[0265] "Means for reproducing thought clones" refers to techniques or methods for reproducing the thought and behavior patterns of a specific person based on learned records.

[0266] "Means of asking questions" are techniques or methods for asking specific questions to the recreated personality and obtaining a response.

[0267] "Means for understanding and verifying behavioral principles" refers to techniques and methods for analyzing a specific person's past behavior and thought patterns, and understanding and verifying those principles.

[0268] "Current information delivery means" are techniques or methods for providing information, such as the latest news articles or research papers, to a particular person's recreated personality.

[0269] "Means of analyzing information" refers to techniques and methods for analyzing given information using natural language processing techniques, etc., and extracting important keywords and context.

[0270] "Means for generating prompt sentences" refers to the technology or method for creating prompt sentences to be input into a generative AI model based on the analysis results.

[0271] "Means for inputting into a generative AI model" refers to the technology or method for inputting the generated prompt sentence into a generative AI model and having it make a prediction.

[0272] "Means for obtaining the prediction results of a generative AI model" refers to the technology or method for obtaining the prediction results obtained from a generative AI model and using them within the system.

[0273] "Means to elicit speech / action" are techniques or methods for eliciting how a specific person's regenerated personality will speak or act in response to events that may occur in the future.

[0274] MODE FOR CARRYING OUT THE INVENTION

[0275] This invention is a system that learns the past actions and thoughts of a specific person and predicts how that person will react by providing the latest information to the reproduced personality. A specific embodiment of this system will be described below.

[0276] Hardware and software used

[0277] This system operates with three main parties: a server, a terminal, and a user. The server receives data, analyzes it, runs the generative AI model, and provides the results. The terminal provides an interface for users to input information and receive results. Users input information about specific individuals, the latest news articles, research papers, etc.

[0278] The server uses the following software and technologies:

[0279] Natural Language Processing (NLP) techniques: Use generative AI models such as Google®'s BERT® and OpenAI's GPT-4.

[0280] Database: A database system for storing entered information.

[0281] Web Server: A web server for receiving input from users and providing results.

[0282] System Operation

[0283] A user uses a terminal to input the latest news articles or research papers into the system. For example, if a user inputs a news article about new developments in quantum computing, the following specific actions occur:

[0284] 1. A user copies and pastes a news article into the input form and clicks the submit button.

[0285] 2. The server receives the text data of the news article and stores it in a database.

[0286] 3. The server analyzes the news article using the BERT model and extracts important keywords and context.

[0287] 4. The server generates a prompt based on the analysis results, for example, "Predict how a particular scientist will react to new advances in quantum computing."

[0288] 5. The prompt sentence generated by the server is input into the generative AI model (GPT-4).

[0289] 6. The server obtains predictions from the generative AI model, such as "This scientist will be very excited about new advances in quantum computing and will believe they will have a major impact on future research."

[0290] 7. The server provides the prediction results to the user, who then checks the results through their device.

[0291] Specific examples

[0292] For example, if a user inputs "a news article about the latest quantum computing advances" into the system, the following prompt will be generated:

[0293] "Predict how a particular scientist will react to a new development in quantum computing."

[0294] The generative AI model (GPT-4) takes this prompt and predicts how a particular scientist will react to the news, for example, "This scientist will be very excited about the new developments in quantum computing and will believe they will have a major impact on future research."

[0295] In this way, the system can predict the reaction of a particular person to the latest information. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0296] Step 1:

[0297] A user inputs the latest news article or research paper.

[0298] Users use their terminal to copy and paste the text of a news article or research paper into the system's input form and click the submit button, which sends the input data in text format to the server.

[0299] Step 2:

[0300] The server receives the entered information.

[0301] The server receives the text data sent by the user and stores it in a database. The input data is stored in its original format.

[0302] Step 3:

[0303] The server analyzes the information.

[0304] The server analyzes the stored text data using natural language processing (NLP) technology. For example, the BERT model is used for this analysis. The server understands the content of the text and extracts important keywords and context. The input is the stored text data, and the output is the keywords and context information that are the analysis results.

[0305] Step 4:

[0306] The server generates a prompt.

[0307] The server generates a prompt based on the analysis results. For example, if the input news article is about new developments in quantum computing, it generates the prompt, "Please predict how a particular scientist will react to the new developments in quantum computing." The input is the analysis results, and the output is the generated prompt.

[0308] Step 5:

[0309] The server inputs a prompt sentence into the generative AI model.

[0310] The server inputs the generated prompt sentence into a generative AI model (e.g., GPT-4). The generative AI model makes a prediction based on the prompt sentence. The input is the generated prompt sentence, and the output is the prediction result by the generative AI model.

[0311] Step 6:

[0312] The server obtains the prediction results of the generative AI model.

[0313] The server obtains a prediction from the generative AI model. For example, it may predict that "This scientist will be very excited about new advances in quantum computing and will believe that they will have a major impact on future research." The input is a prompt to the generative AI model, and the output is the prediction.

[0314] Step 7:

[0315] The server provides the prediction results to the user.

[0316] The server provides the obtained prediction results to the user, who can check the results on the system interface through their terminal. The input is the prediction result from the generative AI model, and the output is the result provided to the user.

[0317] (Application example 3)

[0318] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0319] Conventional systems could recreate a thought clone of a specific person and predict how that person would react to current social conditions and advances in science and technology, but they were unable to predict what type of advertisement a user would respond to based on the user's past behavioral data and interests, and display the most appropriate advertisement. This made it difficult to maximize the effectiveness of advertising.

[0320] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for learning the records of a specific person, means for regenerating a thought clone of the person based on the learned records, means for asking questions to the regenerated personality, means for understanding and verifying the personality's past behavioral principles, means for providing current information to the personality, means for eliciting how the personality will speak / act in response to potential future events, means for collecting the latest information and inputting it into a generative AI model, means for predicting what type of advertisement a user will respond to based on the user's past behavioral data and interests, and means for displaying the most suitable advertisement to the user based on the prediction result. This makes it possible to display the most suitable advertisement to the user.

[0321] A "specific person" refers to a specific individual, and includes information about that individual's intelligence, thoughts, feelings, emotions, morals, ethics, etc.

[0322] A "means of studying records" is a means of studying information that includes a particular person's intellect, thoughts, feelings, emotions, morals, and ethics.

[0323] A "thought clone" is a personality that imitates the thought and behavior patterns of a specific person, recreated based on learned records.

[0324] "Means of asking questions" are means of asking questions to the recreated personality to understand and verify the personality's past behavioral principles.

[0325] A "principle of action" is a principle that indicates the thoughts and judgments that a particular person has based their actions on in the past.

[0326] "Current information" refers to information about current social conditions and advances in science and technology, such as the latest news articles and research papers.

[0327] A "generative AI model" is an artificial intelligence model that generates a specific output based on input information.

[0328] A "prompt" is text data input to a generative AI model that instructs the model to generate a specific output.

[0329] "User past behavioral data" refers to data about the actions and choices a user has made in the past.

[0330] "Interests" is information about areas or topics in which a user is particularly interested.

[0331] "Means for predicting whether a user will respond to an advertisement" refers to a means for predicting what kind of advertisement a user will be interested in based on the user's past behavioral data and interests.

[0332] The "means for displaying the most suitable advertisement" is a means for displaying the most suitable advertisement to the user based on the prediction results.

[0333] A system for implementing this invention studies the records of a specific individual, recreates a thought clone based on those records, and asks questions to the recreated personality to understand and verify that individual's past behavioral principles. It can also provide current information to the personality and elicit how the individual will speak and act in response to potential future events. Furthermore, by collecting the latest information and inputting it into a generative AI model, it can predict what type of advertising a user will respond to based on the user's past behavioral data and interests, and display the most appropriate advertising.

[0334] Hardware and software used

[0335] Hardware: Smartphones, servers

[0336] Software: Python, transformers library, News API

[0337] Data processing and calculation

[0338] 1. Data collection: The server uses the News API to collect the latest news articles, thereby obtaining information on current social situations and scientific and technological advances.

[0339] 2. Initialize the generative AI model: The server uses the transformers library to initialize the GPT-3 model, which is then ready to generate a specific output based on the information it has collected.

[0340] 3. Predicting user response: Based on the user's past behavioral data and interests, the server inputs prompt sentences into the generative AI model to predict how the user will respond to the advertisement.

[0341] 4. Displaying advertisements: The smartphone displays the most suitable advertisements to the user based on the prediction results received from the server.

[0342] Specific examples

[0343] For example, if a user is interested in "technology," the server collects the latest technology-related news and inputs the following prompt sentence into the generative AI model:

[0344] Prompt Sentence Examples

[0345] User Interests: Technology, Sports, Music

[0346] Latest news articles: ['New smartphone released', 'Latest AI technology unveiled']

[0347] Predict what type of ad this user will respond to.

[0348] Based on this prompt, the generative AI model predicts ads for gadgets and software that the user might be interested in and displays them on their smartphone, making it possible to display ads that are optimal for the user.

[0349] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0350] Step 1:

[0351] The server uses the News API to collect the latest news articles. Specifically, it sends a request to the News API to get the latest news article data. The input is the API request, and the output is a list of the latest news articles.

[0352] Step 2:

[0353] The server uses the transformers library to initialize the generative AI model (GPT-3). Specifically, it loads the model and makes it ready for use. The input is the model's initialization parameters, and the output is the initialized generative AI model.

[0354] Step 3:

[0355] The server inputs prompts into the generative AI model based on the user's past behavioral data and interests. Specifically, it generates prompts that combine the user's interests and the latest news articles and inputs them into the model. The inputs are the user's behavioral data and news articles, and the output is the generated prompts.

[0356] Step 4:

[0357] The server inputs a prompt into the generative AI model to predict what kind of ad the user will respond to. Specifically, the server inputs the prompt into the generative AI model and obtains a prediction result from the model. The input is the prompt, and the output is the predicted result of the ad the user will respond to.

[0358] Step 5:

[0359] The server selects the most suitable advertisement for the user based on the prediction results. Specifically, it analyzes the prediction results and selects the advertisement that is most suitable for the user. The input is the prediction results, and the output is the selected advertisement.

[0360] Step 6:

[0361] The device (smartphone) displays the optimal advertisement received from the server to the user. Specifically, it receives advertisement data from the server and displays it on the user's screen. The input is the advertisement data from the server, and the output is the advertisement displayed to the user.

[0362] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0363] "Example 1"

[0364] As one embodiment of the present invention, a system incorporating an emotion engine is provided. This system learns the records of a specific person and reproduces a thought clone of that person. The emotion engine recognizes the person's emotions and provides information that takes into account that person's emotional state. Specifically, when a person shows emotions such as joy or sadness, the system recognizes the emotion and responds accordingly. For example, when a person shows joy, the system responds to the person by sharing the joy. Also, when a person shows sadness, the system provides words of comfort to the person. This allows the system to more accurately understand the person's emotional state and respond more appropriately.

[0365] "Example 2"

[0366] In another embodiment of the present invention, a means for posing questions to the reproduced personality is provided. This means poses questions that take into account the emotional state of the personality. Specifically, when the personality shows emotions such as joy or sadness, the system recognizes the emotion and poses questions accordingly. For example, when the personality shows joy, the system poses a question such as "What made you happy?" to the personality. When the personality shows sadness, the system poses a question such as "What made you sad?" to the personality.

[0367] This allows the system to better understand the emotional state of the person and respond more appropriately.

[0368] "Example 3"

[0369] Furthermore, in another embodiment of the present invention, a means for providing current information to a personality is provided. This means provides information taking into account the personality's emotional state. Specifically, when the personality shows an emotion such as joy or sadness, the emotion is recognized and information is provided according to the emotion. For example, when the personality shows joy, the system provides the personality with information that amplifies the joy. Also, when the personality shows sadness, the system provides the personality with comforting information. This allows the system to more appropriately respond to the personality's emotional state and provide more appropriate information.

[0370] The processing flow of each embodiment will be described below.

[0371] "Example 1"

[0372] Step 1: Learn from records of specific people. In this step, the system learns from records of specific people's actions, words, emotions, etc.

[0373] Step 2: Recreate the person's thought clone based on the learned records. In this step, the system recreates the person's thought patterns based on the learned records.

[0374] Step 3: Recognize the emotion of the reproduced character using the emotion engine. In this step, the system uses the emotion engine to recognize the emotional state of the reproduced character and respond accordingly.

[0375] "Example 2"

[0376] Step 1: Ask questions to the regenerated personality. In this step, the system asks questions to the regenerated personality.

[0377] Step 2: Recognize the emotions of the regenerated personality using the emotion engine. In this step, the system uses the emotion engine to recognize the emotional state of the regenerated personality and asks questions accordingly.

[0378] "Example 3"

[0379] Step 1: Provide current information to the personality. In this step, the system provides current information to the reproduced personality.

[0380] Step 2: Recognize the emotion of the regenerated personality using the emotion engine. In this step, the system uses the emotion engine to recognize the emotional state of the regenerated personality and provide information accordingly.

[0381] Example 1

[0382] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0383] Conventional systems have difficulty reproducing a specific person's thought patterns and behavioral principles, and are particularly lacking in the ability to recognize emotions and respond appropriately based on them. Furthermore, they lack the means to understand past behavioral principles and predict future actions and statements. Therefore, there is a need for a system that can more accurately reproduce a specific person's personality and respond according to their emotions.

[0384] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0385] In this invention, the server includes a means for collecting records of a specific person, a means for preprocessing the collected records, a means for inputting the preprocessed records into a machine learning model to learn the person's thought patterns and principles of behavior, and a means for incorporating an emotion engine to recognize the person's emotions and provide information that takes their emotional state into consideration. This makes it possible to accurately reproduce the specific person's thought patterns and principles of behavior and respond appropriately according to their emotions.

[0386] A "specific person" refers to a specific individual, and is a person whose digital data, such as past actions, statements, copyrighted works, and social media posts, is the target of collection.

[0387] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works such as books and papers, and social media posts.

[0388] "Means of collection" refers to the methods and technologies used to collect digital data of specific individuals, such as using web scraping tools or APIs.

[0389] "Preprocessing means" refers to methods and techniques for processing collected digital data using natural language processing techniques such as cleaning, tokenization, and stemming.

[0390] A "machine learning model" is an artificial intelligence model used to learn a specific person's thought patterns and behavioral principles, and includes models such as GPT-4 and BERT.

[0391] An "emotion engine" refers to a technology or system that recognizes the emotions of a specific person and provides information that takes their emotional state into account.

[0392] A "thinking clone" refers to an artificial intelligence model that reproduces the thought patterns and behavioral principles of a specific person.

[0393] "Means of asking questions" refers to the methods and techniques used to ask questions of the regenerated personality and obtain answers.

[0394] "Principles of behavior" refers to how a particular person has acted in the past and the principles and reasons behind those actions.

[0395] "Means of providing current information" refers to methods and techniques that provide the recreated personality with the latest information and encourage it to think and act based on that information.

[0396] "Means to elicit speech / action" refers to methods and techniques for predicting how the recreated personality will speak or act in response to future events, and achieving those results.

[0397] This invention is a system that collects records of a specific person and reproduces a thinking clone based on those records. This system collects digital data such as the person's past actions, statements, works such as books and papers, and social media posts, and inputs this data into a machine learning model to learn the person's thought patterns and principles of behavior. Furthermore, by incorporating an emotion engine, the system recognizes the person's emotions and provides information that takes their emotional state into account.

[0398] Data collection

[0399] The server uses a web scraping tool or API to collect digital data of a specific person. For example, the server uses an API to collect past tweets of a specific person. Specifically, the server obtains an API key and specifies the user ID of the specific person to obtain tweet data.

[0400] Data Preprocessing

[0401] The server preprocesses the collected data. Specifically, it uses natural language processing techniques such as cleaning the text data (e.g., removing unnecessary HTML tags and special characters), tokenization, and stemming. This processing is performed using libraries. For example, the server tokenizes tweet data and removes unnecessary special characters.

[0402] Input to the machine learning model

[0403] The server inputs the preprocessed data into a machine learning model. This model is designed to learn a specific person's thought patterns and behavioral principles. Models used include generative AI models such as GPT-4 and BERT. For example, the server inputs the preprocessed tweet data into a GPT-4 model to learn a specific person's thought patterns.

[0404] Incorporating an emotion engine

[0405] The server incorporates an emotion engine into the system. This engine recognizes the emotions of a specific person and provides information that takes into account that person's emotional state. Emotion recognition uses an emotion analysis tool (e.g., IBM Watson (registered trademark), Microsoft (registered trademark), Azure (registered trademark)). For example, the server uses IBM Watson Natural Language Understanding to analyze emotions from tweets from a specific person and inputs the results into the emotion engine.

[0406] Thought Clone Regeneration

[0407] The device receives data sent from the server and reproduces a thought clone of a specific person. The emotion engine recognizes the person's emotions and responds accordingly. For example, if a user types "I'm so happy today," the device responds, "That's great! Did something good happen?". Or, if a user types "I'm sad today," the device offers comforting words, "That's terrible. Is there anything I can do to help?"

[0408] Prompt Sentence Examples

[0409] "Collect past tweets from a specific person, train a GPT-4 model using that data, and build a system with an emotion engine. Make the system respond appropriately based on the emotion input by the user."

[0410] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0411] Step 1:

[0412] The server collects digital data of a specific person. As input, it uses the user ID and API key of the specific person. Specifically, the server uses a web scraping tool or API to collect the specific person's past tweets and posts. As output, it obtains the collected digital data.

[0413] Step 2:

[0414] The server preprocesses the collected digital data. It uses the digital data collected in step 1 as input. Specifically, the server cleans the text data (e.g., removes unnecessary HTML tags and special characters), tokenizes, stems, and uses other natural language processing techniques. This process uses libraries. The output is the preprocessed, clean data.

[0415] Step 3:

[0416] The server inputs the preprocessed data into a machine learning model. The data preprocessed in step 2 is used as input. Specifically, the server inputs the data into a generative AI model such as GPT-4 or BERT, which learns the thought patterns and behavioral principles of a specific person. The output is a trained model.

[0417] Step 4:

[0418] The server incorporates the emotion engine into the system. As input, it uses the model trained in step 3 and the emotion data of a specific person. Specifically, the server uses an emotion analysis tool (e.g., IBM Watson or Microsoft Azure) to analyze the emotion of the specific person and inputs the results into the emotion engine. The output is a system with an emotion engine incorporated.

[0419] Step 5:

[0420] The device receives the data sent from the server and reproduces a thought clone of a specific person. As input, it uses the system incorporating the emotion engine obtained in step 4. In concrete terms, the device receives input from the user, and the reproduced thought clone responds appropriately based on that input. For example, if the user inputs "I'm very happy today," the device responds "That's great! Did something good happen?" The output is an appropriate response to the user.

[0421] (Application example 1)

[0422] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0423] While conventional systems can reproduce the thought patterns and behavioral principles of a specific person, it is difficult to provide appropriate information according to the user's emotional state. There is also a need to provide a more personalized experience by recognizing the user's emotions and providing content that corresponds to them.

[0424] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for learning the records of a specific person, means for regenerating a thought clone of the person based on the learned records, means for asking questions to the regenerated personality, means for understanding and verifying the personality's past behavioral principles, means for providing current information to the personality, means for eliciting how the personality will speak / act in response to future events, means for recognizing the user's emotional state, and means for providing information according to the recognized emotional state. This makes it possible to provide personalized information according to the user's emotional state.

[0425] "Means for learning records of a specific person" refers to collecting digital data such as past actions, statements, copyrighted works, and social media posts related to a specific person, and inputting this data into a machine learning model for learning.

[0426] "Means for reproducing a thought clone of a person based on learned records" refers to means for reproducing the thought patterns and principles of behavior of a specific person based on learned data.

[0427] "Means of asking questions to the recreated personality" refers to the means of asking questions to the recreated thought clone and receiving its response.

[0428] "Means to understand and verify the personality's past principles of behavior" refers to a means of analyzing the past actions and statements of the revived thought clone, and understanding and verifying those principles of behavior.

[0429] "Means for providing current information to the personality" refers to a means for providing current information to the recreated thought clone and observing its reaction.

[0430] "Means to draw out how the personality will speak / act in response to future events" is a means of presenting a future scenario to the revived thought clone and drawing out its words and actions in response to that scenario.

[0431] "Means for recognizing a user's emotional state" refers to a means for analyzing emotions from a user's statements, actions, social media posts, etc., and recognizing that emotional state.

[0432] The "means for providing information according to the recognized emotional state" is a means for providing appropriate information or content based on the recognized emotional state of the user.

[0433] The system for implementing this invention has the ability to learn the records of a specific person and reproduce that person's thought clone, as well as recognize the user's emotional state and provide information according to that emotion.

[0434] Hardware and software used

[0435] Hardware

[0436] server

[0437] Smartphone

[0438] head-mounted display

[0439] software

[0440] Python

[0441] Transformers library

[0442] TextBlob Library

[0443] Data processing and calculation

[0444] server

[0445] The server collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts, and inputs this data into a machine learning model for learning. Based on the learned data, the specific person's thought patterns and principles of behavior are reproduced. Questions are asked to the reproduced thinking clone and responses are obtained. The reproduced thinking clone's past actions and statements are analyzed to understand and verify its principles of behavior. Furthermore, the reproduced thinking clone is provided with current information and its reaction is observed. Future scenarios are presented and the model is prompted to make statements and take actions in response to those scenarios.

[0446] Terminal

[0447] The device (smartphone or head-mounted display) analyzes the user's emotions from their statements, actions, and social media posts, and recognizes their emotional state. Based on the recognized emotional state of the user, it provides appropriate information and content.

[0448] Specific examples

[0449] For example, if a user posts on social media, "I'm so tired today," the device analyzes the post and recognizes the user's emotional state as "negative." The server then provides relaxing music or comforting words. Alternatively, if a user posts, "Today was the best day ever!", the device recognizes the post as "positive," and the server provides fun videos or interesting articles.

[0450] Prompt Sentence Examples

[0451] Enter your latest social media posts and the system will recognize the sentiment from them and serve up relevant content.

[0452] In this way, personalized information can be provided according to the user's emotional state.

[0453] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0454] Step 1:

[0455] The server collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. This data serves as input data for learning the specific person's thought patterns and principles of behavior. The collected data is stored in a database.

[0456] Step 2:

[0457] The server inputs the collected data into a machine learning model to learn the thought patterns and behavioral principles of a specific person. This learning process uses the Transformers library to perform natural language processing and extract the characteristics of a specific person. The output is a thought clone of the specific person.

[0458] Step 3:

[0459] The server asks the generated thought clone questions to understand and verify the past behavioral principles of a specific person. Based on the input questions, the thought clone generates a response, which is then output.

[0460] Step 4:

[0461] The server analyzes the past actions and statements of the regenerated thought clone to understand and verify its behavioral principles. This analysis process compares past data with current responses to identify behavioral patterns. The output is an understanding of the behavioral principles.

[0462] Step 5:

[0463] The server provides current information to the regenerated thought clone, allowing the thought clone's reaction to be observed. Current information is given as input, and the thought clone's reaction is obtained as output.

[0464] Step 6:

[0465] The server presents a future scenario and elicits the thinking clone's utterances and actions in response to that scenario. In this process, a future scenario is input and the thinking clone's utterances and actions are output.

[0466] Step 7:

[0467] The device analyzes the user's emotions from their statements, actions, and social media posts to recognize their emotional state. This analysis uses the TextBlob library, which takes the user's latest posts as input and outputs their emotional state.

[0468] Step 8:

[0469] The device provides appropriate information and content based on the user's recognized emotional state. For example, if the user is recognized as "negative," it provides relaxing music and comforting words. The emotional state is given as input, and appropriate content is provided as output.

[0470] In this way, personalized information can be provided according to the user's emotional state.

[0471] Example 2

[0472] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0473] Conventional systems have the problem that they are unable to take into account the emotional state of the recreated personality when asking questions to it, making it difficult to have an appropriate dialogue. Furthermore, they lack the means to predict how the recreated personality will react to modern technology and situations. This makes it difficult for users to have a natural dialogue with the recreated personality.

[0474] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0475] In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of that person based on the learned records, means for asking questions to the reproduced personality, means for analyzing the emotional state of the personality, means for generating questions according to the emotional state, means for understanding and verifying the personality's past principles of behavior, means for providing the personality with current information, and means for eliciting how the personality will speak / act in response to future events. This enables appropriate dialogue that takes into account the emotional state of the reproduced personality, allowing the user to have a natural dialogue with the reproduced personality.

[0476] "Records of a specific person" refers to information about a specific individual, including their intelligence, thoughts, feelings, emotions, morals, and ethics.

[0477] "Means of learning" refers to the process of analyzing records of specific individuals and accumulating knowledge based on that information.

[0478] A "thought clone" refers to a virtual personality that has the intelligence, thoughts, feelings, and emotions of a specific person, recreated based on learned records.

[0479] "Means of regeneration" refers to the process of generating and operating a thought clone based on learned records.

[0480] "Means for asking questions" refers to the process of conveying user-entered questions in an appropriate format to the reproduced personality.

[0481] "Means for analyzing emotional state" refers to the process of analyzing the text data of the reproduced personality to identify its emotional state.

[0482] The "means for generating questions according to emotional state" refers to a process for generating appropriate questions based on the analyzed emotional state.

[0483] "Means of understanding and verifying the principles of past behavior" refers to the process of analyzing the past behavior and thought patterns of the recreated personality and verifying them based on that.

[0484] "Current informational means" refers to the process of providing the recreated personality with information about contemporary technology and situations.

[0485] "Methods of eliciting how to speak / act in response to possible future events" refers to the process of predicting how the recreated personality will react to future events and eliciting that response.

[0486] This invention is a system that learns the records of a specific person and recreates a thought clone based on those records. The system asks questions to the recreated personality, analyzes its emotional state, and generates questions according to that emotional state, allowing the user to have a natural conversation.

[0487] The server uses natural language processing technology to learn the records of a specific person. Specifically, it uses generative AI models such as Google's BERT and OpenAI's GPT-4. This allows it to learn information about that person, including their intelligence, thoughts, emotions, feelings, morals, and ethics.

[0488] Based on the learned records, the server recreates a thought clone, a virtual personality with the intelligence, thoughts, feelings, and emotions of a specific person.

[0489] When a user inputs a question through the device's interface, the device sends the question to the server, which analyzes the received question and converts it into a format appropriate for the reproduced personality. This analysis is performed using a natural language processing model.

[0490] The server uses emotion analysis software to analyze the emotional state of the recreated personality by inputting the text data of the recreated personality and identifying the emotional state.

[0491] Depending on the emotional state, the server generates an appropriate question. For example, if the reproduced personality is happy, the server generates the question "What made you happy?". If the reproduced personality is sad, the server generates the question "What made you sad?".

[0492] The server asks the generated questions to the regenerated personality. This process uses an API that accesses the regenerated personality's database. The regenerated personality generates a response to the received question. This response is generated based on the regenerated personality's database.

[0493] The server transmits the response generated by the reproduced personality to the terminal, and the terminal displays the response received from the server to the user, allowing the user to confirm the response of the reproduced personality.

[0494] As a concrete example, if a user enters the question "What do you think about smartphones, which didn't exist when you were alive?", the following actions will be taken:

[0495] 1. The user types a question into the terminal interface.

[0496] 2. The device sends a question to the server.

[0497] 3. The server parses the question and converts it into a format appropriate for the reproduced personality.

[0498] 4. The server analyzes the emotional state of the recreated personality.

[0499] 5. The server generates questions according to the emotional state.

[0500] 6. The server asks the recreated personality a question.

[0501] 7. The reproduced personality responds, "I think smartphones are very useful tools. They're a technology that was unthinkable in my time."

[0502] 8. The server sends the response to the device.

[0503] 9. The terminal displays the response to the user.

[0504] This allows the user to have a natural conversation with the reproduced personality.

[0505] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0506] Step 1:

[0507] The user enters a question.

[0508] The user inputs a question through the device interface, for example, "What do you think about smartphones, which didn't exist when you were alive?" The input question is temporarily stored in the device's memory.

[0509] Step 2:

[0510] The terminal sends the user's question to the server.

[0511] The terminal sends the question entered by the user to the server using the HTTP protocol. The input data is sent to the server in text format. The server enters a waiting state to analyze the received question.

[0512] Step 3:

[0513] The server analyzes the question and converts it into a format appropriate for the recreated personality.

[0514] The server analyzes the received question using a natural language processing model (for example, OpenAI's GPT-4). The input for the analysis is the user's question, and the output is a question in a format appropriate for the reproduced personality. Specifically, it understands the intent and context of the question and converts it into a format that is easy for the reproduced personality to understand.

[0515] Step 4:

[0516] The server analyzes the emotional state of the recreated personality.

[0517] The server uses emotion analysis software to analyze the emotional state of the reproduced personality. The input is the text data of the reproduced personality, and the output is the emotional state (joy, sadness, anger, etc.). The server analyzes the reproduced personality's past response data to identify the emotional state.

[0518] Step 5:

[0519] The server generates questions according to the emotional state.

[0520] The server generates appropriate questions based on the results of emotion analysis. The input is emotional state data, and the output is a question corresponding to the emotional state. For example, if the reproduced personality shows joy, the server generates the question "What made you happy?". If the reproduced personality shows sadness, the server generates the question "What made you sad?".

[0521] Step 6:

[0522] The server asks the recreated personality a question.

[0523] The server asks the generated questions to the regenerated personality. This process uses an API that accesses the database of regenerated personalities. The input is the generated question, and the output is the regenerated personality's response. The regenerated personality generates a response to the received question.

[0524] Step 7:

[0525] The regenerated personality generates a response.

[0526] The reproduced personality generates a response to the question it receives. This response is generated based on the reproduced personality's database. The input is a question from the server, and the output is the reproduced personality's response. For example, a response such as "I think smartphones are a very useful tool. It's a technology that was unthinkable in my time" can be generated.

[0527] Step 8:

[0528] The server sends a response to the terminal.

[0529] The server sends the response generated by the reproduced personality to the terminal. This communication also uses the HTTP protocol. The input is the response of the reproduced personality, and the output is the data sent to the terminal.

[0530] Step 9:

[0531] The terminal displays the response to the user.

[0532] The terminal displays the response received from the server to the user. The input is the response data from the server, and the output is the text displayed to the user. The user can check the response of the reproduced personality.

[0533] (Application example 2)

[0534] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0535] Previous systems have struggled to ask natural questions to the regenerated personality and to properly analyze its responses. In particular, it has been difficult to ask questions that take into account the regenerated personality's emotional state and to predict its reactions to modern technology. This has limited the user's experience when interacting with the regenerated personality.

[0536] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0537] In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of the person based on the learned records, means for asking questions to the reproduced personality, means for understanding and verifying the personality's past behavioral principles, means for providing the personality with current information, means for eliciting how the personality will speak / act in response to potential future events, means for asking questions that take into account the personality's emotional state, means for generating responses for the personality using a generative AI model, means for generating questions for the personality using prompt sentences, and means for determining emotions based on the personality's responses. This makes it possible to ask natural questions to the reproduced personality and appropriately analyze its responses.

[0538] "Means for studying the records of a specific person" refers to means for collecting information about a specific person and using that information to study that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[0539] "Means for reproducing a thought clone of a person based on learned records" refers to means for reproducing a clone that imitates the thoughts and actions of a specific person based on learned records of that person.

[0540] "Means for asking questions to the regenerated personality" are means for asking natural questions to the regenerated personality.

[0541] "Means for understanding and verifying the principles of past behavior of the personality" refers to means for understanding and verifying the principles of past behavior and thinking of the recreated personality.

[0542] "Means for providing current information to the personality" refers to means for providing modern technology and information to the recreated personality.

[0543] "Means to draw out how the personality will speak / act in response to future events" refers to means for drawing out how the reproduced personality will speak and act in response to future events.

[0544] The "means for asking questions taking into consideration the emotional state of the personality" refers to a means for recognizing the emotional state of the reproduced personality and asking questions in accordance with that state.

[0545] "Means for generating reactions of the personality using a generative AI model" refers to means for generating reactions of the reproduced personality using a generative AI model.

[0546] The "means for generating a question for the character using a prompt sentence" refers to a means for generating a question for the reproduced character using a prompt sentence.

[0547] The "means for determining the emotion based on the reaction of the personality" is a means for analyzing the reaction of the reproduced personality and determining its emotion.

[0548] The following system configuration will be described as an embodiment of the present invention.

[0549] System Configuration

[0550] The system consists of a server and a user terminal. The server has the function of learning the records of a specific person and regenerating a thought clone based on those records. It also has the function of asking questions to the regenerated personality, generating responses, and determining emotions. The user terminal provides an interface for the user to input questions and receive responses from the regenerated personality.

[0551] Hardware and software used

[0552] Hardware: Servers, smartphones

[0553] Software: Python, OpenAI API

[0554] Data processing and calculation

[0555] 1. A means of studying the records of specific individuals

[0556] The server collects information about a specific person and uses that information to learn about that person's intelligence, thoughts, feelings, emotions, morals, and ethics, thereby creating a detailed profile of that person.

[0557] 2. A means of recreating a thought clone of the individual based on the learned records

[0558] The server uses the records of a specific person it has learned to create a clone that mimics that person's thoughts and actions. This clone understands the specific person's past behavioral principles and, by providing them with current information, simulates what they would say and do in response to future events.

[0559] 3. A means of questioning the regenerated personality

[0560] A question entered from the user's device is sent to the server, which then asks the question to the reproduced personality. The question is generated as a prompt sentence, and a generative AI model is used to generate a response from the reproduced personality.

[0561] 4. A method for asking questions that take into account the emotional state of the individual in question

[0562] The server recognizes the emotional state of the recreated personality and generates questions accordingly. For example, if the recreated personality expresses joy, the server will ask a question such as, "What made you happy?"

[0563] 5. Means for generating the personality's responses using a generative AI model

[0564] The server uses a generative AI model to generate reactions for the reproduced personality, which are then sent to the user's device and displayed to the user.

[0565] 6. Means for generating questions for the personality using prompt sentences

[0566] The server uses prompts to generate questions for the persona, such as "What do you think about smartphones, which didn't exist in your lifetime?"

[0567] 7. Means for determining emotions based on the personality's reactions

[0568] The server analyzes the reactions of the reproduced personality and determines its emotions. For example, if the reactions contain keywords such as "happy" or "sad," the server determines the emotions based on those keywords.

[0569] Specific examples

[0570] Consider a case where a user uses a smartphone to ask a question to a specific historical figure. The user inputs the question, "What do you think about smartphones, which didn't exist in your time?" The server generates this question as a prompt sentence and uses a generative AI model to generate a response from the reproduced personality. The generated response might be something like, "Modern technology is very interesting. It would have been unthinkable in my time." The server then analyzes this response and determines its emotion. For example, if the response indicates a positive emotion such as "interesting," the server determines it as "joy."

[0571] Example prompt sentence:

[0572] What do you think about smartphones, which didn't exist in your lifetime?

[0573] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0574] Step 1:

[0575] The user uses the terminal to enter a question.

[0576] Input: A question typed in by the user (e.g., "What do you think about smartphones, which didn't exist in your lifetime?")

[0577] Output: Question text

[0578] Specific actions: The user opens the application on their smartphone, enters a question in the text box for entering a question, and presses the send button.

[0579] Step 2:

[0580] The terminal sends the entered question to the server.

[0581] Input: Question text

[0582] Output: Question data sent to the server

[0583] Specific operation: The terminal sends the entered question text to the server as an HTTP request.

[0584] Step 3:

[0585] The server receives the question and generates a prompt.

[0586] Input: Question data sent to the server

[0587] Output: Prompt statement (e.g., "What do you think about smartphones, which didn't exist in your lifetime?")

[0588] Specific behavior: The server parses the received question data and formats it as a prompt.

[0589] Step 4:

[0590] The server uses a generative AI model to generate the reactions of the played personality.

[0591] Input: prompt statement

[0592] Output: Response of recreated personality (e.g., "Modern technology is very interesting. It was unthinkable in my time.")

[0593] Specific operation: The server calls the OpenAI API, passes the prompt sentence as input to the generative AI model, and receives the generated response.

[0594] Step 5:

[0595] The server analyzes the reactions of the recreated personality and determines its emotions.

[0596] Input: Response from recreated personality

[0597] Output: Emotion determination result (e.g., "joy")

[0598] What happens: The server analyzes the reaction text and determines the sentiment based on specific keywords (e.g., "interesting").

[0599] Step 6:

[0600] The server transmits the generated reaction and emotion determination result to the terminal.

[0601] Input: Generated reaction, emotion judgment result

[0602] Output: Reaction and emotion data sent to the device

[0603] Specific operation: The server sends the generated reaction and emotion judgment result to the terminal as an HTTP response.

[0604] Step 7:

[0605] The terminal displays the received response and emotion determination result to the user.

[0606] Input: Reaction data and emotion data sent to the device

[0607] Output: Reactions displayed to the user and emotion judgment results

[0608] Specific operation: The device analyzes the received data and displays the response text and emotion determination result on the user interface.

[0609] Example 3

[0610] Next, a third embodiment of the third embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0611] In conventional systems, when recreating the thoughts and emotions of a specific person, it was difficult to take into account the latest information and the person's emotional state. As a result, the recreated personality could not respond appropriately to real-world social situations or technological advances, making it impractical. In addition, because information was not provided according to the person's emotional state, the recreated personality's reactions sometimes lacked consistency.

[0612] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0613] In this invention, the server includes means for learning the records of a specific person, means for recreating a thought clone of the person based on the learned records, means for asking questions to the recreated personality, means for understanding and verifying the personality's past behavioral principles, means for providing the personality with current information, means for eliciting how the personality will speak / act in response to potential future events, means for collecting the latest information, means for analyzing the collected information and extracting important keywords and topics, means for generating prompts based on the extracted keywords and topics, means for inputting the generated prompts and predicting how the specific personality will react to the information, and means for recognizing emotional states and providing information corresponding to the recognized emotional states. This allows the recreated personality to respond appropriately to real social situations and technological advances, and to provide consistent information according to emotional states.

[0614] A "record of a specific person" is a collection of information about a specific person, including their intelligence, thoughts, feelings, emotions, morals, and ethics.

[0615] "Means of learning" are methods and techniques for analyzing the records of a particular person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[0616] A "thought clone" is an imitation of a specific person's thoughts and emotions, recreated based on their records.

[0617] "Means of asking questions" are methods or techniques for asking specific questions to the recreated personality and eliciting its response.

[0618] "Past behavioral principles" are basic principles or patterns that indicate how a particular person has thought or acted in the past.

[0619] "Current information" refers to information about current social conditions and technological advances, such as the latest news articles and research papers.

[0620] "Means of collecting the latest information" refers to methods and techniques for collecting the latest news articles and research papers using web scraping tools and APIs.

[0621] "Means of analyzing information" refers to methods and techniques for analyzing collected information using natural language processing tools and extracting important keywords and topics.

[0622] "Means for generating prompt sentences" refers to methods or techniques for creating prompt sentences to be input into generative AI models based on extracted keywords and topics.

[0623] "Means for predicting responses" refers to methods or techniques that input the generated prompt sentences into an AI model and predict how a specific personality will respond to that information.

[0624] "Means for recognizing emotional states" refers to methods or techniques for recognizing the emotional states of the reproduced personalities using emotion analysis tools.

[0625] "Means of providing information" refers to methods or techniques for selecting and providing appropriate information to the reproduced personality according to the recognized emotional state.

[0626] This invention is a system that learns the records of a specific person and reproduces that person's thought clone. The system consists of three main elements: a server, a terminal, and a user.

[0627] The server first collects and studies the records of a specific person. This includes information about the person's intelligence, thoughts, emotions, feelings, morals, and ethics. The server analyzes this information and generates data to recreate a thought clone of the specific person.

[0628] The server then collects information such as the latest news articles and research papers using a web scraping tool or an API (e.g., NewsAPI) and stores the collected information in a database (e.g., MySQL®).

[0629] The server analyzes the collected information using natural language processing (NLP) tools to extract important keywords and topics, and generates prompts based on the analysis results to be input into a generative AI model (e.g., GPT-4).

[0630] The device inputs the generated prompt sentences into the AI ​​model to predict how a specific personality will react to the information, and the prediction results are provided to the user through the device's user interface.

[0631] Furthermore, the server uses emotion analysis tools (e.g., IBM Watson Tone Analyzer or Microsoft Azure) to recognize the emotional state of the person. Depending on the emotional state, the server selects appropriate information and provides it to the user through the terminal.

[0632] As a concrete example, consider the case where the latest news item is "A vaccine for the new coronavirus has been developed." The server analyzes this information and generates the following prompt:

[0633] "How do you feel when you hear the news that a vaccine for COVID-19 has been developed?"

[0634] The device then inputs this prompt into an AI model to predict how a particular personality will respond—for example, the AI ​​model might predict "feel joy."

[0635] Furthermore, using emotion analysis tools, if the personality is actually expressing joy, the server will provide information that amplifies the joy, such as "The successful development of a COVID-19 vaccine has increased the chances of saving people around the world."

[0636] In this way, the system can provide appropriate information according to the emotional state of the personality. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0637] Step 1:

[0638] The server collects and learns records of specific people.

[0639] Input: Information containing a specific person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[0640] Specific operation: The server retrieves data about a specific person from the database and analyzes it using a machine learning algorithm, thereby generating data for recreating a thought clone of the specific person.

[0641] Output: Data to recreate a thought clone of a specific person.

[0642] Step 2:

[0643] The server collects information such as the latest news articles and research papers.

[0644] Input: URL of news site or academic paper site.

[0645] Specific operation: The server uses a web scraping tool or API (e.g., NewsAPI) to collect the latest information from the specified site. The collected information is stored in a database.

[0646] Output: Collected news articles and research paper data.

[0647] Step 3:

[0648] The server analyzes the collected information and extracts important keywords and topics.

[0649] Input: Data collected from news articles and research papers.

[0650] How it works: The server uses natural language processing (NLP) tools to analyze the collected information, extracting frequently occurring keywords and topics.

[0651] Output: Extracted keywords and topics.

[0652] Step 4:

[0653] The server generates a prompt sentence based on the extracted keywords and topics.

[0654] Input: Extracted keywords or topics.

[0655] Specific operation: The server creates a prompt to be input to a generative AI model (e.g., GPT-4). For example, in response to the information, "A vaccine for the new coronavirus has been developed," the server generates the prompt, "How do you feel when you hear the news that a vaccine for the new coronavirus has been developed?"

[0656] Output: The generated prompt statement.

[0657] Step 5:

[0658] The device inputs the generated prompts into an AI model that predicts how a particular personality will react to the information.

[0659] Input: The generated prompt statement.

[0660] Specific operation: The device inputs the prompt sentence received from the server into the AI ​​model, which then predicts the response of a specific personality based on the input prompt sentence.

[0661] Output: Predicted response.

[0662] Step 6:

[0663] The server uses an emotion analysis tool to recognize the emotional state of the persona.

[0664] Input: predicted response.

[0665] Specific operation: The server uses an emotion analysis tool (e.g., IBM Watson Tone Analyzer or Microsoft Azure) to analyze the predicted reaction and recognize the emotional state.

[0666] Output: Perceived emotional state.

[0667] Step 7:

[0668] The server selects appropriate information according to the recognized emotional state and provides it to the user through the terminal.

[0669] Input: Perceived emotional state.

[0670] Specific operation: The server searches the database for and selects information that corresponds to the emotional state. For example, if the personality expresses joy, the server selects information that amplifies joy, such as "The successful development of a COVID-19 vaccine has increased the chances of saving people around the world." The selected information is sent to the device and displayed to the user.

[0671] Output: Pertinent information provided to the user.

[0672] (Application example 3)

[0673] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0674] Conventional systems were capable of reproducing thought clones based on the records of specific individuals, and understanding and verifying past behavioral principles, but lacked the ability to recognize the user's emotional state in real time and provide information accordingly. This meant that they were unable to provide appropriate information based on the user's emotions, resulting in a poor user experience. Furthermore, the lack of information provided based on the user's emotional state made it difficult to provide services that were sensitive to the user's emotions.

[0675] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0676] In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of the person based on the learned records, means for asking questions to the reproduced personality, means for understanding and verifying the personality's past principles of behavior, means for providing current information to the personality, means for eliciting how the personality will speak / act in response to future events, means for recognizing the user's emotional state, and means for providing information based on the recognized emotional state, thereby making it possible to provide appropriate information according to the user's emotional state.

[0677] A "means for learning records of a specific person" is a device or program that collects information about a specific person's intelligence, thoughts, feelings, emotions, morals, ethics, etc., and learns based on this information.

[0678] "Means for reproducing a thought clone of a person based on learned records" refers to a device or program that generates a clone that imitates the thoughts and actions of a specific person based on learned records of that person.

[0679] "Means for asking questions to the regenerated personality" is a device or program for asking questions to the regenerated thought clone and obtaining its response.

[0680] "Means for understanding and verifying the personality's past behavioral principles" refers to a device or program that analyzes the past behavior and thought patterns of the reproduced thought clone, and understands and verifies those principles.

[0681] A "means for providing current information to said personality" is a device or program for providing current information to the recreated thought clone, such as the latest news articles or research papers.

[0682] "Means to derive how the personality will speak / act in response to future events" refers to a device or program that predicts how the regenerated thought clone will speak and act in response to future events and derives the results.

[0683] "Means for recognizing the user's emotional state" refers to a device or program that analyzes the user's facial expressions, voice, actions, etc., and recognizes their emotional state in real time.

[0684] A "means for providing information based on a recognized emotional state" is a device or program for providing appropriate news articles or content in response to a recognized emotional state of a user.

[0685] The system for implementing this invention has the ability to learn the records of a specific person and reproduce thought clones based on those records, as well as recognize the user's emotional state in real time and provide information according to that emotion.

[0686] Hardware and software used

[0687] Hardware: Smartphones, servers

[0688] Software: Python, EmotionRecognizer, NewsAPI

[0689] System configuration

[0690] 1. Means of studying a particular person's record:

[0691] The server collects information on a specific person's intelligence, thoughts, emotions, feelings, morals, and ethics, and uses this information to learn, thereby understanding the behavior and thought patterns of that specific person.

[0692] 2. A means of recreating a thought clone of the individual based on the learned records:

[0693] The server generates a clone that mimics the thoughts and actions of a specific person based on the learned records of that person. This clone can reproduce the specific person's past behavioral principles.

[0694] 3. Means of questioning the regenerated personality:

[0695] The server asks questions to the regenerated thought clone and receives its responses. These questions are used to understand and verify the past behavioral principles of a specific person.

[0696] 4. Means of understanding and verifying the individual's past motivations:

[0697] The server analyzes the past actions and thought patterns of the recreated thought clone, and understands and verifies its principles, which allows for a deep understanding of the principles of a specific person's behavior.

[0698] 5. Means of providing current information to the individual:

[0699] The server provides the regenerated thought clone with current information, such as the latest news articles and research papers, allowing it to predict how the thought clone will react to current social conditions and advances in science and technology.

[0700] 6. Methods to elicit how the individual will respond to future events:

[0701] The server predicts how the regenerated thought clone will speak and act in response to future events and derives the results.

[0702] 7. Means of recognizing the user's emotional state:

[0703] The smartphone analyzes the user's facial expressions, voice, and behavior to recognize their emotional state in real time. It uses EmotionRecognizer to accurately grasp the user's emotions.

[0704] 8. Means of providing information based on perceived emotional states:

[0705] The server provides appropriate news articles and content according to the user's recognized emotional state. It analyzes the emotions of news articles and filters information that matches the user's emotions.

[0706] Specific examples

[0707] Emotion recognition is performed through the camera and microphone while the user is using the smartphone.

[0708] If the user is feeling "joyed," positive news articles (e.g., "New technology is changing lives") will be displayed.

[0709] If the user is feeling "sad," they will see comforting and encouraging news articles (e.g., "Community overcomes difficulties").

[0710] Prompt Sentence Examples

[0711] "Generate positive news articles to show users when they're feeling joy."

[0712] "Generate comforting news articles to show users when they're feeling sad."

[0713] In this way, a personalized news feed can be provided according to the user's emotional state.

[0714] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0715] Step 1:

[0716] The server collects records of specific individuals. As input, it receives information about the individual's intelligence, thoughts, emotions, feelings, morals, and ethics. Based on this data, it learns the individual's behavior and thought patterns. As output, it generates a record of the learned individual.

[0717] Step 2:

[0718] The server reproduces a thought clone of a specific person based on the learned record. As input, it receives the learned record generated in step 1. Based on this, it generates a clone that imitates the thoughts and actions of the specific person. As output, it obtains the reproduced thought clone.

[0719] Step 3:

[0720] The server asks questions to the reproduced thought clone. As input, it receives the reproduced thought clone and the question. This gives a response to understand and verify the past behavioral principles of a specific person. As output, the response of the thought clone is generated.

[0721] Step 4:

[0722] The server understands and verifies the past behavioral principles of the reproduced thought clone. As input, it receives the response obtained in step 3. It analyzes this and processes the data to understand the behavioral principles of a specific person. As output, it obtains the understanding results of the behavioral principles.

[0723] Step 5:

[0724] The server provides current information, such as the latest news articles and research papers, to the regenerated thought clone. As input, it receives the current information and the regenerated thought clone. This predicts how the thought clone will react to current social conditions and advances in science and technology. As output, it obtains the thought clone's reaction.

[0725] Step 6:

[0726] The server predicts how the regenerated thought clone will speak and act in response to future events, and derives the results. As input, it receives the reaction of the thought clone obtained in step 5. Based on this, it predicts what the thought clone will say and do in response to future events. The predicted result is obtained as output.

[0727] Step 7:

[0728] The device recognizes the user's emotional state in real time. As input, it receives data such as the user's facial expressions, voice, and behavior. It uses EmotionRecognizer to analyze this data and recognize the user's emotional state. As output, it obtains the recognized emotional state.

[0729] Step 8:

[0730] The server provides appropriate news articles and content according to the recognized emotional state of the user. As input, it receives the emotional state obtained in step 7 and news article data. It analyzes the emotions of the news articles and filters information that matches the user's emotions. As output, it obtains a personalized news feed.

[0731] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0732] The data generation model 58 is a so-called generative AI (Artificial Intelligence).

[0733] An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generative AIs. Data generation models 58

[0734] , obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in the data format of voice data, text data, etc. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0735] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0736] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0737] [Second embodiment]

[0738] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0739] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0740] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0741] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0742] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0743] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0744] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0745] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0746] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is performed by the processor 28 executing the specific processing program 56 on the RAM 30.

[0747] This is realized by operating as a specific processing unit 290 in accordance with the specific processing program 56 executed above.

[0748] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0749] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0750] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0751] "Example 1"

[0752] In one embodiment of the present invention, a machine learning model is developed to learn the records of a specific individual by collecting digital data such as the individual's past actions, statements, books, papers, and social media posts. This data is then input into a machine learning model that learns the individual's thought patterns and principles of behavior, enabling the reproduction of that individual's personality.

[0753] "Example 2"

[0754] As a means of asking questions to the recreated personality, natural language processing technology is used to provide an interface that allows humans to ask questions naturally. For example, by asking a question such as, "What do you think about smartphones, which didn't exist in your lifetime?", it is possible to predict how the person will react to modern technology.

[0755] "Example 3"

[0756] To provide current information to the personality, information such as the latest news articles and research papers is input into the system, which can then predict how the person will react to current social conditions and advances in science and technology.

[0757] The processing flow of each embodiment will be described below.

[0758] "Example 1"

[0759] Step 1: Collect digital data such as a specific person's past actions, statements, works such as books and papers, and social media posts.

[0760] Step 2: Preprocess the collected data and convert it into a format suitable for the machine learning model.

[0761] Step 3: The preprocessed data is input into the machine learning model and the model is trained. As a result of the training, a model that has learned a person's thought patterns and principles of behavior is obtained.

[0762] "Example 2"

[0763] Step 1: Using natural language processing technology, develop an interface that allows humans to ask questions naturally.

[0764] Step 2: Ask the recreated personality questions through the developed interface. Step 3: Analyze the responses from the recreated personality and predict how the person will react to modern technology.

[0765] "Example 3"

[0766] Step 1: Gather information such as the latest news articles and research papers.

[0767] Step 2: Enter the collected information into the system.

[0768] Step 3: Based on the input information, predict how the recreated personality will react to current social conditions and advances in science and technology.

[0769] Example 1

[0770] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0771] Conventional systems have had difficulty accurately reproducing a specific person's thought patterns and principles of behavior. Furthermore, preprocessing of collected data and building, training, and evaluating machine learning models were often done manually, resulting in inefficiencies. Furthermore, the process of posing appropriate questions to the reproduced personality and generating responses was complicated.

[0772] The identification process by the identification processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means. In this invention, the server includes a means for collecting digital data and performing preprocessing, a means for constructing and training a machine learning model, a means for evaluating the trained model, and a means for inputting a prompt sentence into the generative AI model and generating a response. This makes it possible to efficiently and accurately reproduce the thought patterns and principles of behavior of a specific person, ask appropriate questions to the reproduced personality, and generate a response.

[0773] A "specific person" refers to a specific individual, and is the subject of collection of digital data such as that individual's past actions, statements, copyrighted works, and social media posts.

[0774] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0775] "Thought clone" refers to a machine learning model built to replicate a specific person's thought patterns and behavioral principles.

[0776] "Digital data" refers to electronically stored information such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0777] "Preprocessing" refers to the process of cleaning, tokenizing, normalizing, and otherwise converting collected digital data into a format suitable for machine learning models.

[0778] A "machine learning model" refers to an algorithm or neural network used to learn and replicate a specific person's thought patterns and behavioral principles.

[0779] "Learning" refers to the process of inputting data into a machine learning model so that the model can understand and reproduce a specific person's thought patterns and principles of behavior.

[0780] "Evaluation" refers to the testing and verification process to confirm the predictive accuracy and performance of a trained machine learning model.

[0781] "Generative AI model" refers to an artificial intelligence model that is used to reproduce a specific person's thought patterns and principles of behavior and generate responses to prompts.

[0782] A "prompt" refers to a question or instruction entered into a generative AI model, and is input data used to generate a response based on a specific person's thought patterns.

[0783] This invention is a system that learns the records of a specific person and reproduces that person's thought patterns and principles of behavior. A specific embodiment of this system will be described below.

[0784] First, the user collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. This data serves as the basis for learning the person's thought patterns and principles of behavior. The collected data is then sent to a server.

[0785] The server then preprocesses the collected digital data, specifically cleaning, tokenizing, and normalizing the text data. The software used includes Python libraries, such as removing unnecessary HTML tags from social media posts and splitting the text into words and sentences.

[0786] Using the preprocessed data, the server builds a machine learning model. This model is used to learn a specific person's thought patterns and behavioral principles. Specifically, it uses a deep learning framework such as TensorFlow. Building the model includes dividing the dataset, selecting a model, and designing a neural network.

[0787] The server then inputs the preprocessed data into the machine learning model it has built and trains the model. The model is trained using a training set and an optimization algorithm such as gradient descent is used to minimize the loss function.

[0788] Once the training is complete, the server evaluates the model. The evaluation checks the model's predictive accuracy using a test set. Evaluation metrics include precision, recall, and F1 score.

[0789] Finally, the user uses the trained model to recreate a specific person's personality. Specifically, the user inputs a prompt into the generative AI model, which simulates how the specific person would respond. For example, the user can input the following prompt:

[0790] "Tell me about your latest book."

[0791] The generative AI model recreates the thought patterns of a specific person based on the input prompt and generates a response that is highly realistic because it is based on the specific person's past behavioral principles and thought patterns.

[0792] The above is a specific embodiment for carrying out the present invention. This system makes it possible to efficiently and accurately reproduce the thought patterns and behavioral principles of a specific person, pose appropriate questions to the reproduced personality, and generate responses.

[0793] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0794] Step 1: Data collection

[0795] A user collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. The input is digital data collected from the internet or databases. The output is the collected digital data sent to a server. Specifically, a web scraping tool is used to automatically collect social media posts.

[0796] Step 2: Data Preprocessing

[0797] The server preprocesses the collected digital data. The input is the collected digital data. The output is the preprocessed data. Specific operations include cleaning, tokenizing, and normalizing the text data. For example, a Python library is used to tokenize the text and remove unnecessary HTML tags.

[0798] Step 3: Building a machine learning model

[0799] The server builds a machine learning model using the preprocessed data. The input is the preprocessed data. The output is the built machine learning model. Specifically, it designs a neural network model using a deep learning framework such as TensorFlow. It splits the dataset into a training set and a test set, and selects and designs a model.

[0800] Step 4: Training the model

[0801] The server inputs the preprocessed data into the constructed machine learning model and trains the model. The inputs are the preprocessed data and the constructed model. The output is a trained model. Specifically, the model is trained using a training set, and an optimization algorithm such as gradient descent is used to minimize the loss function.

[0802] Step 5: Evaluate the model

[0803] The server evaluates the model after training. The inputs are the trained model and a test set. The output is the model's predictive accuracy and performance. Specifically, it checks the model's predictive accuracy using the test set and calculates evaluation metrics such as precision, recall, and F1 score.

[0804] Step 6: Personality Regeneration

[0805] The user uses a trained model to recreate the personality of a specific person. The inputs are the trained model and a prompt. The output is a generated response. Specifically, the prompt is input into the generative AI model, and a simulation is performed to see how the specific person would respond. For example, the prompt is input as "Tell me about your latest book," and the model generates a response.

[0806] (Application example 1)

[0807] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0808] In today's world, there is a demand for systems that can reproduce the knowledge and thought patterns of specific famous people or historical figures, allowing users to converse with them. However, with conventional technology, it is difficult to effectively collect digital data about a specific person, input it into a machine learning model, and generate answers to user questions that are appropriate for that person. Therefore, the challenge is to provide a system that can accurately reproduce the thought patterns and principles of behavior of a specific person, allowing users to converse with that person.

[0809] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0810] In this invention, the server includes a means for learning the records of a specific person, a means for reproducing a thought clone of the person based on the learned records, and a means for asking questions to the reproduced personality. This makes it possible to collect digital data of a specific person, input it into a machine learning model for learning, and generate answers to user questions that are characteristic of that person.

[0811] "Specific person" refers to a specific individual, such as a particular celebrity or historical figure.

[0812] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0813] A "thinking clone" refers to an artificial intelligence model that reproduces the thought patterns and behavioral principles of a specific person.

[0814] A "machine learning model" refers to an algorithm or system that learns a specific person's thought patterns and behavioral principles based on collected digital data.

[0815] "Digital data" refers to electronically stored information such as a specific person's past actions, statements, copyrighted works, and social media posts.

[0816] "User" means an individual who uses the System to interact with a specific person.

[0817] "Questions" refer to questions that the user poses to the recreated personality.

[0818] "Answer" refers to the response that the recreated personality generates to the user's question.

[0819] A "thinking pattern" refers to how a particular person tends to think and judge a particular situation or problem.

[0820] A "principle of action" refers to a basic principle of how a particular person should act in a particular situation or problem.

[0821] To implement this invention, it is necessary to build a system that collects digital data of specific individuals and inputs it into a machine learning model for learning. This system consists of three main elements: a server, a terminal, and a user.

[0822] server

[0823] The server includes a means for learning the records of a specific person, a means for reproducing a thought clone of that person based on the learned records, and a means for asking questions to the reproduced personality. Specifically, the server performs the following processes.

[0824] 1. Data collection: The server collects digital data such as social media posts, copyrighted works, and statements of specific individuals. This data is collected automatically using APIs.

[0825] 2. Data input: Convert the collected data into JSON format and input it into a machine learning model such as OpenAI's GPT-3.

[0826] 3. Model learning: The GPT-3 model learns a specific person's thought patterns and behavioral principles based on the input data.

[0827] 4. Response generation: The trained model is used to generate a response that is appropriate for the user's question.

[0828] Terminal

[0829] A terminal provides an interface for users to interact with the system. Devices such as smartphones and tablets fall into this category. Terminals have the following functions:

[0830] 1. User Interface: Provides an interface for users to enter questions and receive answers.

[0831] 2. Data transmission: The question entered by the user is sent to the server.

[0832] 3. Data reception: Receives the response sent from the server and displays it to the user.

[0833] User

[0834] A user is an individual who uses the system to interact with a specific person. The user performs the following operations:

[0835] 1. Question input: The user inputs a question for a specific person through the terminal interface.

[0836] 2. Receiving the answer: The user receives the answer sent from the server on their terminal and checks it.

[0837] Specific examples

[0838] For example, if a user selects "famous scientists" as a specific person, the user can enter a question such as:

[0839] "Tell me about your next project."

[0840] "Tell me about the theory of relativity."

[0841] The server uses the trained model to generate answers to these questions and provides them to the user via the terminal.

[0842] Prompt Sentence Examples

[0843] "I discovered something new today."

[0844] I'm thinking about my next project.

[0845] "Science is constantly evolving."

[0846] In this way, the user can experience the knowledge and thought patterns of a particular person and enjoy the dialogue.

[0847] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0848] Step 1:

[0849] The server collects digital data of a specific person. Specifically, it automatically collects data such as social media posts, copyrighted works, and statements using an API. The input is the person's identifying information, and the output is the collected digital data.

[0850] Step 2:

[0851] The server converts the collected digital data into JSON format. The input is the collected digital data, and the output is JSON format data. This conversion makes the data suitable for machine learning models.

[0852] Step 3:

[0853] The server inputs the JSON-formatted digital data into a machine learning model such as OpenAI's GPT-3, training the model. The input is JSON-formatted data, and the output is a trained model. The model learns a specific person's thought patterns and principles of behavior.

[0854] Step 4:

[0855] A user inputs a question for a specific person through the terminal interface. The input is the user's question, and the output is the question data sent from the terminal to the server.

[0856] Step 5:

[0857] The terminal sends the question entered by the user to the server. The input is the user's question data, and the output is the data to be sent to the server.

[0858] Step 6:

[0859] The server generates answers to user questions using the trained model. The input is the user's question data and the trained model, and the output is the generated answer. The server inputs the question as a prompt to the model and generates an answer.

[0860] Step 7:

[0861] The server sends the generated answer to the terminal. The input is the generated answer, and the output is the data sent to the terminal.

[0862] Step 8:

[0863] The terminal receives the response sent from the server and displays it to the user. The input is the data sent from the server, and the output is the response displayed to the user.

[0864] In this way, the user can experience the knowledge and thought patterns of a particular person and enjoy the dialogue.

[0865] Example 2

[0866] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0867] In conventional systems, the process of asking questions to the reproduced personality and receiving responses was complicated, making it difficult for users to ask questions naturally. Also, it was difficult to predict how the reproduced personality would respond to modern technology and information.

[0868] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a question, a means for a terminal to send a question to the server, a means for the server to analyze the question, a means for the server to generate a prompt sentence for the generative AI model, a means for the server to input a prompt sentence to the generative AI model, a means for the server to receive a generated response, a means for the server to send the response to the terminal, and a means for the terminal to display the response to the user. This allows the user to naturally ask a question and easily obtain a response from the reproduced personality to the question.

[0869] "Records of a specific person" are data that include information about a specific person's intelligence, thoughts, feelings, emotions, morals, ethics, etc.

[0870] "Means of learning" are methods and techniques for analyzing the records of a particular person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[0871] "Means for reproducing thought clones" refers to methods and techniques for generating artificial intelligence that imitates the thoughts and actions of a specific person based on the records of that person that it has learned.

[0872] A "means for asking a question" is a method or technique that allows a user to naturally input a question to the reproduced personality and transmit the question to the system.

[0873] "Means for understanding and verifying the principles of past behavior" refers to methods and techniques for analyzing the past behavior and thought patterns of the recreated personality and predicting responses to current questions based on that analysis.

[0874] "Current means of providing information" are methods and techniques for providing modern technology and information to the recreated personality and eliciting its response.

[0875] "Means of eliciting how to speak / act in response to possible future events" refers to methods or techniques for predicting how the recreated personality will react to future events and generating that response.

[0876] The "means for the user to input a question" is an interface that allows the user to input a question to the reproduced personality using a terminal.

[0877] The "means by which the terminal transmits the question to the server" refers to the communication protocol or technology used to transmit the question entered by the user from the terminal to the server.

[0878] The "means by which the server analyzes the question" refers to a method or technology by which the server analyzes the question received by using natural language processing technology and extracts the grammatical structure and keywords of the question.

[0879] "Means for generating prompt sentences for a generative AI model" refers to methods or technologies for generating appropriate prompt sentences for a generative AI model based on the analysis results.

[0880] "Means for inputting prompt sentences into a generative AI model" refers to methods or techniques for inputting generated prompt sentences into a generative AI model and obtaining a response from the model.

[0881] "Means for receiving the generated response" refers to the method or technology by which the server receives the response from the generative AI model.

[0882] The "means for sending the response to the terminal" refers to the communications protocol or technology by which the server sends the generated response to the user's terminal.

[0883] The "means by which the terminal displays the response to the user" is an interface for displaying the response received by the terminal to the user.

[0884] This invention is a system for posing questions to a recreated personality and obtaining responses. The system includes means for learning a record of a specific person and recreating a thought clone based on the record. The system also includes means for posing questions to the recreated personality and generating responses to the questions.

[0885] Hardware and software used

[0886] Hardware: Servers, user devices (PCs, smartphones, etc.)

[0887] Software: Natural language processing libraries, generative AI models (e.g., GPT-4)

[0888] Specific operation of the system

[0889] 1. User enters a question:

[0890] The user uses the device to input questions for the recreated personality, such as "What do you think about smartphones, which didn't exist in your lifetime?"

[0891] 2. The device sends a question to the server:

[0892] The terminal sends the question entered by the user to the server using a communication protocol such as an HTTP request.

[0893] 3. The server parses the question:

[0894] The server analyzes the received question using a natural language processing library. Specifically, it analyzes the grammatical structure of the question and extracts keywords. For example, keywords such as "smartphone" and "What do you think?" are extracted.

[0895] 4. The server generates a prompt for the generative AI model:

[0896] Based on the analysis results, the server generates a prompt for the generative AI model (e.g., GPT-4). For example, it generates a prompt such as, "You are a person from the 19th century. What do you think about modern smartphones?"

[0897] 5. The server inputs a prompt to the generative AI model:

[0898] The server inputs the generated prompt sentence into the generative AI model, which then generates a response based on the prompt sentence.

[0899] 6. The server receives the generated response:

[0900] The server receives a response from the generative AI model, such as, "I think smartphones are very useful tools, but they were unimaginable in my day."

[0901] 7. The server sends the response to the device:

[0902] The server then sends the generated response to the user's device, again using a communication protocol such as an HTTP response.

[0903] 8. The terminal displays the response to the user:

[0904] The terminal displays the received response to the user, who can then check the response from the personality played back on the screen.

[0905] Specific examples

[0906] User question: "What do you think about smartphones, which didn't exist in your lifetime?"

[0907] Prompt for generative AI model: "You are a person from the 19th century. What do you think about modern smartphones?"

[0908] This system allows users to naturally ask questions to the reproduced personality and easily obtain responses to those questions.

[0909] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0910] Step 1:

[0911] The user enters a question.

[0912] The user inputs a question using the device's keyboard or touch screen. Once input is complete, the user clicks the "Submit" button. The input data is a text-based question. For example, a question might be entered such as, "What do you think about smartphones, which didn't exist when you were alive?"

[0913] Step 2:

[0914] The terminal sends a question to the server.

[0915] The device converts the input question into JSON format and sends it to the server as an HTTP POST request. The input data is the question in text format, and the output data is the question data in JSON format.

[0916] Step 3:

[0917] The server parses the question.

[0918] The server parses the received JSON data and extracts the question. It uses a natural language processing library to analyze the grammatical structure of the question and extract keywords. The input data is the question data in JSON format, and the output data is the analysis result (grammatical structure and keywords). For example, keywords such as "smartphone" and "What do you think?" are extracted.

[0919] Step 4:

[0920] The server generates a prompt for the generative AI model.

[0921] Based on the analysis results, the server generates a prompt for the generative AI model (e.g., GPT-4). The input data is the analysis results, and the output data is the prompt. For example, it generates a prompt such as, "You are a person from the 19th century. What do you think about modern smartphones?"

[0922] Step 5:

[0923] The server inputs a prompt sentence into the generative AI model.

[0924] The server inputs the generated prompt sentence into the generative AI model and waits for the model's response. The input data is the prompt sentence, and the output data is the response from the generative AI model.

[0925] Step 6:

[0926] The server receives the generated response.

[0927] The server receives the response from the generative AI model and checks the response. The input data is the response from the generative AI model, and the output data is the checked response. For example, a response such as "I think smartphones are very useful tools, but they were unimaginable in my day" is generated.

[0928] Step 7:

[0929] The server sends a response to the terminal.

[0930] The server converts the generated response back into JSON format and sends it to the user's device as an HTTP response. The input data is the confirmed response, and the output data is the JSON-formatted response data.

[0931] Step 8:

[0932] The terminal displays the response to the user.

[0933] The device parses the received JSON data and displays the response on the screen. The input data is the response data in JSON format, and the output data is the response in text format. The user can check the displayed response and enjoy interacting with the reproduced personality.

[0934] (Application example 2)

[0935] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0936] Previous systems had issues with the interactive experience, as it was difficult for users to ask questions naturally to the reproduced personality and there was no function to play back the answers aloud. Furthermore, the user experience was limited because there was no way for users to ask questions by voice or text input.

[0937] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of the person based on the learned records, means for posing questions to the reproduced personality, means for understanding and verifying the personality's past behavioral principles, means for providing the personality with current information, means for eliciting how the personality will speak / act in response to potential future events, means for reproducing the generated answers aloud using speech synthesis technology, and means for the user to pose questions by voice input or text input. This allows the user to naturally pose questions and reproduce the generated answers aloud, enabling a more interactive and richer experience.

[0938] "Records of a specific person" refers to information about a specific person, including their intelligence, thoughts, feelings, emotions, morals, and ethics.

[0939] "Means of learning" refers to techniques for analyzing records of a specific person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[0940] A "thought clone" refers to an artificial intelligence that imitates the intelligence, thoughts, emotions, feelings, morals and ethics of a specific person based on learned records.

[0941] "Means for asking questions" refers to an interface that allows the user to naturally ask questions to the recreated personality.

[0942] "Past behavioral principles" refers to principles or patterns that show how a particular person has thought or acted in the past.

[0943] "Current information" refers to data about modern technology and events.

[0944] "Means of eliciting what to say / do" refers to technology that predicts how the recreated personality will react to future events and provides that reaction to the user.

[0945] "Speech synthesis technology" refers to the technology that converts text data into speech.

[0946] "Means for playing back by voice" refers to technology for providing the generated answer to the user by voice using speech synthesis technology.

[0947] "Voice or text input means for asking questions" refers to an interface that allows a user to ask questions to the played-out persona using voice or text.

[0948] The system for implementing this invention includes a program for learning the records of a specific person and recreating a thought clone based on the records. The system allows the user to naturally ask questions to the regenerated personality and provides a function for replaying the answers by voice.

[0949] Hardware and software used

[0950] Hardware: Smartphone

[0951] Software: Python, OpenAI API (GPT-3), speech synthesis library

[0952] Data processing and calculation

[0953] 1. User input: The user uses their smartphone to ask a question by voice or text input.

[0954] 2. Question processing: The question is sent to the server, and an answer is generated for the specified personality using OpenAI's GPT-3 model.

[0955] 3. Answer Generation: The GPT-3 model generates appropriate answers to questions based on records of a specific person.

[0956] 4. Speech synthesis: The generated answer is converted into speech using a library.

[0957] 5. Audio playback: The answer is played aloud to the user through the smartphone speaker.

[0958] Specific examples

[0959] A user opens the app and asks a question to a specific historical figure. For example, "What do you think about smartphones, which didn't exist in your time?" This question is sent to a server, and the GPT-3 model generates an answer for that person. The answer might be something like, "Smartphones allow instant access to information and will greatly contribute to scientific advancement." This answer is then played aloud using speech synthesis technology and provided to the user.

[0960] Prompt Sentence Examples

[0961] As a specific historical figure, answer the following question: What do you think about smartphones, which didn't exist in your lifetime?

[0962] In this way, users can naturally ask questions and have the generated answers played back to them aloud, resulting in a more interactive and enriching experience.

[0963] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0964] Step 1:

[0965] A user starts the application using a smartphone and selects a specific historical figure. Next, the user inputs a question by voice or text. This input data is text data containing the user's question.

[0966] Step 2:

[0967] The device sends the question entered by the user to the server. The server analyzes the received question and sends a prompt sentence to the generative AI model (GPT-3) to generate an answer as a specific historical figure. This prompt sentence is text data containing the user's question.

[0968] Step 3:

[0969] The server receives a response from the generative AI model (GPT-3), which generates an answer to the question based on the records of specific historical figures. This generated answer is returned to the server as text data.

[0970] Step 4:

[0971] The server uses a speech synthesis library (pyttsx3) to convert the received text data into speech data. The speech synthesis library analyzes the text data and generates the corresponding speech data.

[0972] Step 5:

[0973] The server transmits the generated voice data to the device, which then plays the received voice data and provides it to the user. The user can then listen to the audio of the specific historical figure's response through the smartphone speaker.

[0974] Step 6:

[0975] The user can then enter the question again or select a different historical figure, allowing the user to continue the interactive experience.

[0976] Example 3

[0977] Next, a description will be given of Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0978] While conventional systems can reproduce a specific person's past actions and thoughts, it is difficult to predict how that person will react to the latest information. Furthermore, there is a lack of concrete means for inputting information such as the latest news articles or research papers and predicting how a specific person will react to that information. This makes it difficult to accurately predict a specific person's future actions and statements.

[0979] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0980] In this invention, the server includes means for learning the records of a specific person, means for recreating a thought clone of the person based on the learned records, means for asking questions to the recreated personality, means for understanding and verifying the personality's past behavioral principles, means for providing current information to the personality, means for analyzing the information, means for generating prompt sentences based on the analysis results, means for inputting the generated prompt sentences into a generative AI model, means for obtaining prediction results from the generative AI model, and means for deriving how the personality will speak / act in response to possible future events. This makes it possible to predict how a specific person will react to the latest information.

[0981] A "specific person" is an individual person that the system targets, and is the subject of learning records of that person's past actions, thoughts, emotions, etc.

[0982] "Means of studying records" refers to techniques and methods for collecting information, including a specific person's intelligence, thoughts, feelings, emotions, morals, and ethics, and for analyzing and studying this information.

[0983] "Means for reproducing thought clones" refers to techniques or methods for reproducing the thought and behavior patterns of a specific person based on learned records.

[0984] "Means of asking questions" are techniques or methods for asking specific questions to the recreated personality and obtaining a response.

[0985] "Means for understanding and verifying behavioral principles" refers to techniques and methods for analyzing a specific person's past behavior and thought patterns, and understanding and verifying those principles.

[0986] "Current information delivery means" are techniques or methods for providing information, such as the latest news articles or research papers, to a particular person's recreated personality.

[0987] "Means of analyzing information" refers to techniques and methods for analyzing given information using natural language processing techniques, etc., and extracting important keywords and context.

[0988] "Means for generating prompt sentences" refers to the technology or method for creating prompt sentences to be input into a generative AI model based on the analysis results.

[0989] "Means for inputting into a generative AI model" refers to the technology or method for inputting the generated prompt sentence into a generative AI model and having it make a prediction.

[0990] "Means for obtaining the prediction results of a generative AI model" refers to the technology or method for obtaining the prediction results obtained from a generative AI model and using them within the system.

[0991] "Means to elicit speech / action" are techniques or methods for eliciting how a specific person's regenerated personality will speak or act in response to events that may occur in the future.

[0992] MODE FOR CARRYING OUT THE INVENTION

[0993] This invention is a system that learns the past actions and thoughts of a specific person and predicts how that person will react by providing the latest information to the reproduced personality. A specific embodiment of this system will be described below.

[0994] Hardware and software used

[0995] This system operates with three main parties: a server, a terminal, and a user. The server receives data, analyzes it, runs the generative AI model, and provides the results. The terminal provides an interface for users to input information and receive results. Users input information about specific individuals, the latest news articles, research papers, etc.

[0996] The server uses the following software and technologies:

[0997] Natural Language Processing (NLP) techniques: Using generative AI models such as Google's BERT and OpenAI's GPT-4.

[0998] Database: A database system for storing entered information.

[0999] Web Server: A web server for receiving input from users and providing results.

[1000] System Operation

[1001] A user uses a terminal to input the latest news articles or research papers into the system. For example, if a user inputs a news article about new developments in quantum computing, the following specific actions occur:

[1002] 1. A user copies and pastes a news article into the input form and clicks the submit button.

[1003] 2. The server receives the text data of the news article and stores it in a database.

[1004] 3. The server analyzes the news article using the BERT model and extracts important keywords and context.

[1005] 4. The server generates a prompt based on the analysis results, for example, "Predict how a particular scientist will react to new advances in quantum computing."

[1006] 5. The prompt sentence generated by the server is input into the generative AI model (GPT-4).

[1007] 6. The server obtains predictions from the generative AI model, such as "This scientist will be very excited about new advances in quantum computing and will believe they will have a major impact on future research."

[1008] 7. The server provides the prediction results to the user, who then checks the results through their device.

[1009] Specific examples

[1010] For example, if a user inputs "a news article about the latest quantum computing advances" into the system, the following prompt will be generated:

[1011] "Predict how a particular scientist will react to a new development in quantum computing."

[1012] The generative AI model (GPT-4) takes this prompt and predicts how a particular scientist will react to the news, for example, "This scientist will be very excited about the new developments in quantum computing and will believe they will have a major impact on future research."

[1013] In this way, the system can predict the reaction of a particular person to the latest information. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1014] Step 1:

[1015] A user inputs the latest news article or research paper.

[1016] Users use their terminal to copy and paste the text of a news article or research paper into the system's input form and click the submit button, which sends the input data in text format to the server.

[1017] Step 2:

[1018] The server receives the entered information.

[1019] The server receives the text data sent by the user and stores it in a database. The input data is stored in its original format.

[1020] Step 3:

[1021] The server analyzes the information.

[1022] The server analyzes the stored text data using natural language processing (NLP) technology. For example, the BERT model is used for this analysis. The server understands the content of the text and extracts important keywords and context. The input is the stored text data, and the output is the keywords and context information that are the analysis results.

[1023] Step 4:

[1024] The server generates a prompt.

[1025] The server generates a prompt based on the analysis results. For example, if the input news article is about new developments in quantum computing, it generates the prompt, "Please predict how a particular scientist will react to the new developments in quantum computing." The input is the analysis results, and the output is the generated prompt.

[1026] Step 5:

[1027] The server inputs a prompt sentence into the generative AI model.

[1028] The server inputs the generated prompt sentence into a generative AI model (e.g., GPT-4). The generative AI model makes a prediction based on the prompt sentence. The input is the generated prompt sentence, and the output is the prediction result by the generative AI model.

[1029] Step 6:

[1030] The server obtains the prediction results of the generative AI model.

[1031] The server obtains a prediction from the generative AI model. For example, it may predict that "This scientist will be very excited about new advances in quantum computing and will believe that they will have a major impact on future research." The input is a prompt to the generative AI model, and the output is the prediction.

[1032] Step 7:

[1033] The server provides the prediction results to the user.

[1034] The server provides the obtained prediction results to the user, who can check the results on the system interface through their terminal. The input is the prediction result from the generative AI model, and the output is the result provided to the user.

[1035] (Application example 3)

[1036] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1037] Conventional systems could recreate a thought clone of a specific person and predict how that person would react to current social conditions and advances in science and technology, but they were unable to predict what type of advertisement a user would respond to based on the user's past behavioral data and interests, and display the most appropriate advertisement. This made it difficult to maximize the effectiveness of advertising.

[1038] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for learning the records of a specific person, means for regenerating a thought clone of the person based on the learned records, means for asking questions to the regenerated personality, means for understanding and verifying the personality's past behavioral principles, means for providing current information to the personality, means for eliciting how the personality will speak / act in response to potential future events, means for collecting the latest information and inputting it into a generative AI model, means for predicting what type of advertisement a user will respond to based on the user's past behavioral data and interests, and means for displaying the most suitable advertisement to the user based on the prediction result. This makes it possible to display the most suitable advertisement to the user.

[1039] A "specific person" refers to a specific individual, and includes information about that individual's intelligence, thoughts, feelings, emotions, morals, ethics, etc.

[1040] A "means of studying records" is a means of studying information that includes a particular person's intellect, thoughts, feelings, emotions, morals, and ethics.

[1041] A "thought clone" is a personality that imitates the thought and behavior patterns of a specific person, recreated based on learned records.

[1042] "Means of asking questions" are means of asking questions to the recreated personality to understand and verify the personality's past behavioral principles.

[1043] A "principle of action" is a principle that indicates the thoughts and judgments that a particular person has based their actions on in the past.

[1044] "Current information" refers to information about current social conditions and advances in science and technology, such as the latest news articles and research papers.

[1045] A "generative AI model" is an artificial intelligence model that generates a specific output based on input information.

[1046] A "prompt" is text data input to a generative AI model that instructs the model to generate a specific output.

[1047] "User past behavioral data" refers to data about the actions and choices a user has made in the past.

[1048] "Interests" is information about areas or topics in which a user is particularly interested.

[1049] "Means for predicting whether a user will respond to an advertisement" refers to a means for predicting what kind of advertisement a user will be interested in based on the user's past behavioral data and interests.

[1050] The "means for displaying the most suitable advertisement" is a means for displaying the most suitable advertisement to the user based on the prediction results.

[1051] A system for implementing this invention studies the records of a specific individual, recreates a thought clone based on those records, and asks questions to the recreated personality to understand and verify that individual's past behavioral principles. It can also provide current information to the personality and elicit how the individual will speak and act in response to potential future events. Furthermore, by collecting the latest information and inputting it into a generative AI model, it can predict what type of advertising a user will respond to based on the user's past behavioral data and interests, and display the most appropriate advertising.

[1052] Hardware and software used

[1053] Hardware: Smartphones, servers

[1054] Software: Python, transformers library, News API

[1055] Data processing and calculation

[1056] 1. Data collection: The server uses the News API to collect the latest news articles, thereby obtaining information on current social situations and scientific and technological advances.

[1057] 2. Initialize the generative AI model: The server uses the transformers library to initialize the GPT-3 model, which is then ready to generate a specific output based on the information it has collected.

[1058] 3. Predicting user response: Based on the user's past behavioral data and interests, the server inputs prompt sentences into the generative AI model to predict how the user will respond to the advertisement.

[1059] 4. Displaying advertisements: The smartphone displays the most suitable advertisements to the user based on the prediction results received from the server.

[1060] Specific examples

[1061] For example, if a user is interested in "technology," the server collects the latest technology-related news and inputs the following prompt sentence into the generative AI model:

[1062] Prompt Sentence Examples

[1063] User Interests: Technology, Sports, Music

[1064] Latest news articles: ['New smartphone released', 'Latest AI technology unveiled']

[1065] Predict what type of ad this user will respond to.

[1066] Based on this prompt, the generative AI model predicts ads for gadgets and software that the user might be interested in and displays them on their smartphone, making it possible to display ads that are optimal for the user.

[1067] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1068] Step 1:

[1069] The server uses the News API to collect the latest news articles. Specifically, it sends a request to the News API to get the latest news article data. The input is the API request, and the output is a list of the latest news articles.

[1070] Step 2:

[1071] The server uses the transformers library to initialize the generative AI model (GPT-3). Specifically, it loads the model and makes it ready for use. The input is the model's initialization parameters, and the output is the initialized generative AI model.

[1072] Step 3:

[1073] The server inputs prompts into the generative AI model based on the user's past behavioral data and interests. Specifically, it generates prompts that combine the user's interests and the latest news articles and inputs them into the model. The inputs are the user's behavioral data and news articles, and the output is the generated prompts.

[1074] Step 4:

[1075] The server inputs a prompt into the generative AI model to predict what kind of ad the user will respond to. Specifically, the server inputs the prompt into the generative AI model and obtains a prediction result from the model. The input is the prompt, and the output is the predicted result of the ad the user will respond to.

[1076] Step 5:

[1077] The server selects the most suitable advertisement for the user based on the prediction results. Specifically, it analyzes the prediction results and selects the advertisement that is most suitable for the user. The input is the prediction results, and the output is the selected advertisement.

[1078] Step 6:

[1079] The device (smartphone) displays the optimal advertisement received from the server to the user. Specifically, it receives advertisement data from the server and displays it on the user's screen. The input is the advertisement data from the server, and the output is the advertisement displayed to the user.

[1080] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1081] "Example 1"

[1082] As one embodiment of the present invention, a system incorporating an emotion engine is provided. This system learns the records of a specific person and reproduces a thought clone of that person. The emotion engine recognizes the person's emotions and provides information that takes into account that person's emotional state. Specifically, when a person shows emotions such as joy or sadness, the system recognizes the emotion and responds accordingly. For example, when a person shows joy, the system responds to the person by sharing the joy. Also, when a person shows sadness, the system provides words of comfort to the person. This allows the system to more accurately understand the person's emotional state and respond more appropriately.

[1083] "Example 2"

[1084] In another embodiment of the present invention, a means for posing questions to the reproduced personality is provided. This means poses questions that take into account the emotional state of the personality. Specifically, when the personality shows emotions such as joy or sadness, the system recognizes the emotion and poses questions accordingly. For example, when the personality shows joy, the system poses a question such as "What made you happy?" to the personality. When the personality shows sadness, the system poses a question such as "What made you sad?" to the personality. This allows the system to more deeply understand the emotional state of the personality and respond more appropriately.

[1085] "Example 3"

[1086] Furthermore, in another embodiment of the present invention, a means for providing current information to a personality is provided. This means provides information taking into account the personality's emotional state. Specifically, when the personality shows an emotion such as joy or sadness, the emotion is recognized and information is provided according to the emotion. For example, when the personality shows joy, the system provides the personality with information that amplifies the joy. Also, when the personality shows sadness, the system provides the personality with comforting information. This allows the system to more appropriately respond to the personality's emotional state and provide more appropriate information.

[1087] The processing flow of each embodiment will be described below.

[1088] "Example 1"

[1089] Step 1: Learn from records of specific people. In this step, the system learns from records of specific people's actions, words, emotions, etc.

[1090] Step 2: Recreate the person's thought clone based on the learned records. In this step, the system recreates the person's thought patterns based on the learned records.

[1091] Step 3: Recognize the emotion of the reproduced character using the emotion engine. In this step, the system uses the emotion engine to recognize the emotional state of the reproduced character and respond accordingly.

[1092] "Example 2"

[1093] Step 1: Ask questions to the regenerated personality. In this step, the system asks questions to the regenerated personality.

[1094] Step 2: Recognize the emotions of the regenerated personality using the emotion engine. In this step, the system uses the emotion engine to recognize the emotional state of the regenerated personality and asks questions accordingly.

[1095] "Example 3"

[1096] Step 1: Present current information to the personality. In this step, the system replays the current information.

[1097] Give it to the living personality.

[1098] Step 2: Recognize the emotion of the regenerated personality using the emotion engine. In this step, the system uses the emotion engine to recognize the emotional state of the regenerated personality and provide information accordingly.

[1099] Example 1

[1100] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1101] Conventional systems have difficulty reproducing a specific person's thought patterns and behavioral principles, and are particularly lacking in the ability to recognize emotions and respond appropriately based on them. Furthermore, they lack the means to understand past behavioral principles and predict future actions and statements. Therefore, there is a need for a system that can more accurately reproduce a specific person's personality and respond according to their emotions.

[1102] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1103] In this invention, the server includes a means for collecting records of a specific person, a means for preprocessing the collected records, a means for inputting the preprocessed records into a machine learning model to learn the person's thought patterns and principles of behavior, and a means for incorporating an emotion engine to recognize the person's emotions and provide information that takes their emotional state into consideration. This makes it possible to accurately reproduce the specific person's thought patterns and principles of behavior and respond appropriately according to their emotions.

[1104] A "specific person" refers to a specific individual, and is a person whose digital data, such as past actions, statements, copyrighted works, and social media posts, is the target of collection.

[1105] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works such as books and papers, and social media posts.

[1106] "Means of collection" refers to the methods and technologies used to collect digital data of specific individuals, such as using web scraping tools or APIs.

[1107] "Preprocessing means" refers to methods and techniques for processing collected digital data using natural language processing techniques such as cleaning, tokenization, and stemming.

[1108] A "machine learning model" is an artificial intelligence model used to learn a specific person's thought patterns and behavioral principles, and includes models such as GPT-4 and BERT.

[1109] An "emotion engine" refers to a technology or system that recognizes the emotions of a specific person and provides information that takes their emotional state into account.

[1110] A "thinking clone" refers to an artificial intelligence model that reproduces the thought patterns and behavioral principles of a specific person.

[1111] "Means of asking questions" refers to the methods and techniques used to ask questions of the regenerated personality and obtain answers.

[1112] "Principles of behavior" refers to how a particular person has acted in the past and the principles and reasons behind those actions.

[1113] "Means of providing current information" refers to methods and techniques that provide the recreated personality with the latest information and encourage it to think and act based on that information.

[1114] "Means to elicit speech / action" refers to methods and techniques for predicting how the recreated personality will speak or act in response to future events, and achieving those results.

[1115] This invention is a system that collects records of a specific person and reproduces a thinking clone based on those records. This system collects digital data such as the person's past actions, statements, works such as books and papers, and social media posts, and inputs this data into a machine learning model to learn the person's thought patterns and principles of behavior. Furthermore, by incorporating an emotion engine, the system recognizes the person's emotions and provides information that takes their emotional state into account.

[1116] Data collection

[1117] The server uses a web scraping tool or API to collect digital data of a specific person. For example, the server uses an API to collect past tweets of a specific person. Specifically, the server obtains an API key and specifies the user ID of the specific person to obtain tweet data.

[1118] Data Preprocessing

[1119] The server preprocesses the collected data. Specifically, it uses natural language processing techniques such as cleaning the text data (e.g., removing unnecessary HTML tags and special characters), tokenization, and stemming. This processing is performed using libraries. For example, the server tokenizes tweet data and removes unnecessary special characters.

[1120] Input to the machine learning model

[1121] The server inputs the preprocessed data into a machine learning model. This model is designed to learn a specific person's thought patterns and behavioral principles. Models used include generative AI models such as GPT-4 and BERT. For example, the server inputs the preprocessed tweet data into a GPT-4 model to learn a specific person's thought patterns.

[1122] Incorporating an emotion engine

[1123] The server incorporates an emotion engine into the system. This engine recognizes the emotions of a specific person and provides information that takes their emotional state into consideration. Emotion recognition uses an emotion analysis tool (e.g., IBM Watson Microsoft Azure). For example, the server uses IBM Watson Natural Language Understanding to analyze emotions from tweets from a specific person and inputs the results into the emotion engine.

[1124] Thought Clone Regeneration

[1125] The device receives data sent from the server and reproduces a thought clone of a specific person. The emotion engine recognizes the person's emotions and responds accordingly. For example, if a user types "I'm so happy today," the device responds, "That's great! Did something good happen?". Or, if a user types "I'm sad today," the device offers comforting words, "That's terrible. Is there anything I can do to help?"

[1126] Prompt Sentence Examples

[1127] "Collect past tweets from a specific person, train a GPT-4 model using that data, and build a system with an emotion engine. Make the system respond appropriately based on the emotion input by the user."

[1128] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1129] Step 1:

[1130] The server collects digital data of a specific person. As input, it uses the user ID and API key of the specific person. Specifically, the server uses a web scraping tool or API to collect the specific person's past tweets and posts. As output, it obtains the collected digital data.

[1131] Step 2:

[1132] The server preprocesses the collected digital data. It uses the digital data collected in step 1 as input. Specifically, the server cleans the text data (e.g., removes unnecessary HTML tags and special characters), tokenizes, stems, and uses other natural language processing techniques. This process uses libraries. The output is the preprocessed, clean data.

[1133] Step 3:

[1134] The server inputs the preprocessed data into a machine learning model. The data preprocessed in step 2 is used as input. Specifically, the server inputs the data into a generative AI model such as GPT-4 or BERT, which learns the thought patterns and behavioral principles of a specific person. The output is a trained model.

[1135] Step 4:

[1136] The server incorporates the emotion engine into the system. As input, it uses the model trained in step 3 and the emotion data of a specific person. Specifically, the server uses an emotion analysis tool (e.g., IBM Watson or Microsoft Azure) to analyze the emotion of the specific person and inputs the results into the emotion engine. The output is a system with an emotion engine incorporated.

[1137] Step 5:

[1138] The device receives the data sent from the server and reproduces a thought clone of a specific person. As input, it uses the system incorporating the emotion engine obtained in step 4. In concrete terms, the device receives input from the user, and the reproduced thought clone responds appropriately based on that input. For example, if the user inputs "I'm very happy today," the device responds "That's great! Did something good happen?" The output is an appropriate response to the user.

[1139] (Application example 1)

[1140] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1141] While conventional systems can reproduce the thought patterns and behavioral principles of a specific person, it is difficult to provide appropriate information according to the user's emotional state. There is also a need to provide a more personalized experience by recognizing the user's emotions and providing content that corresponds to them.

[1142] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for learning the records of a specific person, means for regenerating a thought clone of the person based on the learned records, means for asking questions to the regenerated personality, means for understanding and verifying the personality's past behavioral principles, means for providing current information to the personality, means for eliciting how the personality will speak / act in response to future events, means for recognizing the user's emotional state, and means for providing information according to the recognized emotional state. This makes it possible to provide personalized information according to the user's emotional state.

[1143] "Means for learning records of a specific person" refers to collecting digital data such as past actions, statements, copyrighted works, and social media posts related to a specific person, and inputting this data into a machine learning model for learning.

[1144] "Means for reproducing a thought clone of a person based on learned records" refers to means for reproducing the thought patterns and principles of behavior of a specific person based on learned data.

[1145] "Means of asking questions to the recreated personality" refers to the means of asking questions to the recreated thought clone and receiving its response.

[1146] "Means to understand and verify the personality's past principles of behavior" refers to a means of analyzing the past actions and statements of the revived thought clone, and understanding and verifying those principles of behavior.

[1147] "Means for providing current information to the personality" refers to a means for providing current information to the recreated thought clone and observing its reaction.

[1148] "Means to draw out how the personality will speak / act in response to future events" is a means of presenting a future scenario to the revived thought clone and drawing out its words and actions in response to that scenario.

[1149] "Means for recognizing a user's emotional state" refers to a means for analyzing emotions from a user's statements, actions, social media posts, etc., and recognizing that emotional state.

[1150] The "means for providing information according to the recognized emotional state" is a means for providing appropriate information or content based on the recognized emotional state of the user.

[1151] The system for implementing this invention has the ability to learn the records of a specific person and reproduce that person's thought clone, as well as recognize the user's emotional state and provide information according to that emotion.

[1152] Hardware and software used

[1153] Hardware

[1154] server

[1155] Smartphone

[1156] head-mounted display

[1157] software

[1158] Python

[1159] Transformers library

[1160] TextBlob Library

[1161] Data processing and calculation

[1162] server

[1163] The server collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts, and inputs this data into a machine learning model for learning. Based on the learned data, the specific person's thought patterns and principles of behavior are reproduced. Questions are asked to the reproduced thinking clone and responses are obtained. The reproduced thinking clone's past actions and statements are analyzed to understand and verify its principles of behavior. Furthermore, the reproduced thinking clone is provided with current information and its reaction is observed. Future scenarios are presented and the model is prompted to make statements and take actions in response to those scenarios.

[1164] Terminal

[1165] The device (smartphone or head-mounted display) analyzes the user's emotions from their statements, actions, and social media posts, and recognizes their emotional state. Based on the recognized emotional state of the user, it provides appropriate information and content.

[1166] Specific examples

[1167] For example, if a user posts on social media, "I'm so tired today," the device analyzes the post and recognizes the user's emotional state as "negative." The server then provides relaxing music or comforting words. Alternatively, if a user posts, "Today was the best day ever!", the device recognizes the post as "positive," and the server provides fun videos or interesting articles.

[1168] Prompt Sentence Examples

[1169] Enter your latest social media posts and the system will recognize the sentiment from them and serve up relevant content.

[1170] In this way, personalized information can be provided according to the user's emotional state.

[1171] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1172] Step 1:

[1173] The server collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. This data serves as input data for learning the specific person's thought patterns and principles of behavior. The collected data is stored in a database.

[1174] Step 2:

[1175] The server inputs the collected data into a machine learning model to learn the thought patterns and behavioral principles of a specific person. This learning process uses the Transformers library to perform natural language processing and extract the characteristics of a specific person. The output is a thought clone of the specific person.

[1176] Step 3:

[1177] The server asks the generated thought clone questions to understand and verify the past behavioral principles of a specific person. Based on the input questions, the thought clone generates a response, which is then output.

[1178] Step 4:

[1179] The server analyzes the past actions and statements of the regenerated thought clone to understand and verify its behavioral principles. This analysis process compares past data with current responses to identify behavioral patterns. The output is an understanding of the behavioral principles.

[1180] Step 5:

[1181] The server provides current information to the regenerated thought clone, allowing the thought clone's reaction to be observed. Current information is given as input, and the thought clone's reaction is obtained as output.

[1182] Step 6:

[1183] The server presents a future scenario and elicits the thinking clone's utterances and actions in response to that scenario. In this process, a future scenario is input and the thinking clone's utterances and actions are output.

[1184] Step 7:

[1185] The device analyzes the user's emotions from their statements, actions, and social media posts to recognize their emotional state. This analysis uses the TextBlob library, which takes the user's latest posts as input and outputs their emotional state.

[1186] Step 8:

[1187] The device provides appropriate information and content based on the user's recognized emotional state. For example, if the user is recognized as "negative," it provides relaxing music and comforting words. The emotional state is given as input, and appropriate content is provided as output.

[1188] In this way, personalized information can be provided according to the user's emotional state.

[1189] Example 2

[1190] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1191] Conventional systems have the problem that they are unable to take into account the emotional state of the recreated personality when asking questions to it, making it difficult to have an appropriate dialogue. Furthermore, they lack the means to predict how the recreated personality will react to modern technology and situations. This makes it difficult for users to have a natural dialogue with the recreated personality.

[1192] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1193] In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of that person based on the learned records, means for asking questions to the reproduced personality, means for analyzing the emotional state of the personality, means for generating questions according to the emotional state, means for understanding and verifying the personality's past principles of behavior, means for providing the personality with current information, and means for eliciting how the personality will speak / act in response to future events. This enables appropriate dialogue that takes into account the emotional state of the reproduced personality, allowing the user to have a natural dialogue with the reproduced personality.

[1194] "Records of a specific person" refers to information about a specific individual, including their intelligence, thoughts, feelings, emotions, morals, and ethics.

[1195] "Means of learning" refers to the process of analyzing records of specific individuals and accumulating knowledge based on that information.

[1196] A "thought clone" refers to a virtual personality that has the intelligence, thoughts, feelings, and emotions of a specific person, recreated based on learned records.

[1197] "Means of regeneration" refers to the process of generating and operating a thought clone based on learned records.

[1198] "Means for asking questions" refers to the process of conveying user-entered questions in an appropriate format to the reproduced personality.

[1199] "Means for analyzing emotional state" refers to the process of analyzing the text data of the reproduced personality to identify its emotional state.

[1200] The "means for generating questions according to emotional state" refers to a process for generating appropriate questions based on the analyzed emotional state.

[1201] "Means of understanding and verifying the principles of past behavior" refers to the process of analyzing the past behavior and thought patterns of the recreated personality and verifying them based on that.

[1202] "Current informational means" refers to the process of providing the recreated personality with information about contemporary technology and situations.

[1203] "Methods of eliciting how to speak / act in response to possible future events" refers to the process of predicting how the recreated personality will react to future events and eliciting that response.

[1204] This invention is a system that learns the records of a specific person and recreates a thought clone based on those records. The system asks questions to the recreated personality, analyzes its emotional state, and generates questions according to that emotional state, allowing the user to have a natural conversation.

[1205] The server uses natural language processing technology to learn the records of a specific person. Specifically, it uses generative AI models such as Google's BERT and OpenAI's GPT-4. This allows it to learn information about that person, including their intelligence, thoughts, emotions, feelings, morals, and ethics.

[1206] Based on the learned records, the server recreates a thought clone, a virtual personality with the intelligence, thoughts, feelings, and emotions of a specific person.

[1207] When a user inputs a question through the device's interface, the device sends the question to the server, which analyzes the received question and converts it into a format appropriate for the reproduced personality. This analysis is performed using a natural language processing model.

[1208] The server uses emotion analysis software to analyze the emotional state of the recreated personality by inputting the text data of the recreated personality and identifying the emotional state.

[1209] Depending on the emotional state, the server generates an appropriate question. For example, if the reproduced personality is happy, the server generates the question "What made you happy?". If the reproduced personality is sad, the server generates the question "What made you sad?".

[1210] The server asks the generated questions to the regenerated personality. This process uses an API that accesses the regenerated personality's database. The regenerated personality generates a response to the received question. This response is generated based on the regenerated personality's database.

[1211] The server transmits the response generated by the reproduced personality to the terminal, and the terminal displays the response received from the server to the user, allowing the user to check the response of the reproduced personality.

[1212] As a concrete example, if a user enters the question "What do you think about smartphones, which didn't exist when you were alive?", the following actions will be taken:

[1213] 1. The user types a question into the terminal interface.

[1214] 2. The device sends a question to the server.

[1215] 3. The server parses the question and converts it into a format appropriate for the reproduced personality.

[1216] 4. The server analyzes the emotional state of the recreated personality.

[1217] 5. The server generates questions according to the emotional state.

[1218] 6. The server asks the recreated personality a question.

[1219] 7. The reproduced personality responds, "I think smartphones are very useful tools. They're a technology that was unthinkable in my time."

[1220] 8. The server sends the response to the device.

[1221] 9. The terminal displays the response to the user.

[1222] This allows the user to have a natural conversation with the reproduced personality.

[1223] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1224] Step 1:

[1225] The user enters a question.

[1226] The user inputs a question through the device interface, for example, "What do you think about smartphones, which didn't exist when you were alive?" The input question is temporarily stored in the device's memory.

[1227] Step 2:

[1228] The terminal sends the user's question to the server.

[1229] The terminal sends the question entered by the user to the server using the HTTP protocol. The input data is sent to the server in text format. The server enters a waiting state to analyze the received question.

[1230] Step 3:

[1231] The server analyzes the question and converts it into a format appropriate for the recreated personality.

[1232] The server analyzes the received question using a natural language processing model (for example, OpenAI's GPT-4). The input for the analysis is the user's question, and the output is a question in a format appropriate for the reproduced personality. Specifically, it understands the intent and context of the question and converts it into a format that is easy for the reproduced personality to understand.

[1233] Step 4:

[1234] The server analyzes the emotional state of the recreated personality.

[1235] The server uses emotion analysis software to analyze the emotional state of the reproduced personality. The input is the text data of the reproduced personality, and the output is the emotional state (joy, sadness, anger, etc.). The server analyzes the reproduced personality's past response data to identify the emotional state.

[1236] Step 5:

[1237] The server generates questions according to the emotional state.

[1238] The server generates appropriate questions based on the results of emotion analysis. The input is emotional state data, and the output is a question corresponding to the emotional state. For example, if the reproduced personality shows joy, the server generates the question "What made you happy?". If the reproduced personality shows sadness, the server generates the question "What made you sad?".

[1239] Step 6:

[1240] The server asks the recreated personality a question.

[1241] The server asks the generated questions to the regenerated personality. This process uses an API that accesses the database of regenerated personalities. The input is the generated question, and the output is the regenerated personality's response. The regenerated personality generates a response to the received question.

[1242] Step 7:

[1243] The regenerated personality generates a response.

[1244] The reproduced personality generates a response to the question it receives. This response is generated based on the reproduced personality's database. The input is a question from the server, and the output is the reproduced personality's response. For example, a response such as "I think smartphones are a very useful tool. It's a technology that was unthinkable in my time" can be generated.

[1245] Step 8:

[1246] The server sends a response to the terminal.

[1247] The server sends the response generated by the reproduced personality to the terminal. This communication also uses the HTTP protocol. The input is the response of the reproduced personality, and the output is the data sent to the terminal.

[1248] Step 9:

[1249] The terminal displays the response to the user.

[1250] The terminal displays the response received from the server to the user. The input is the response data from the server, and the output is the text displayed to the user. The user can check the response of the reproduced personality.

[1251] (Application example 2)

[1252] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1253] Previous systems have struggled to ask natural questions to the regenerated personality and to properly analyze its responses. In particular, it has been difficult to ask questions that take into account the regenerated personality's emotional state and to predict its reactions to modern technology. This has limited the user's experience when interacting with the regenerated personality.

[1254] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1255] In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of the person based on the learned records, means for asking questions to the reproduced personality, means for understanding and verifying the personality's past behavioral principles, means for providing the personality with current information, means for eliciting how the personality will speak / act in response to potential future events, means for asking questions that take into account the personality's emotional state, means for generating responses for the personality using a generative AI model, means for generating questions for the personality using prompt sentences, and means for determining emotions based on the personality's responses. This makes it possible to ask natural questions to the reproduced personality and appropriately analyze its responses.

[1256] "Means for studying the records of a specific person" refers to means for collecting information about a specific person and using that information to study that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[1257] "Means for reproducing a thought clone of a person based on learned records" refers to means for reproducing a clone that imitates the thoughts and actions of a specific person based on learned records of that person.

[1258] "Means for asking questions to the regenerated personality" are means for asking natural questions to the regenerated personality.

[1259] "Means for understanding and verifying the principles of the personality's past behavior" refers to means for understanding and verifying the principles of the past behavior and thinking of the recreated personality.

[1260] "Means for providing current information to the personality" refers to means for providing modern technology and information to the recreated personality.

[1261] "Means to draw out how the personality will speak / act in response to future events" is a means to draw out how the reproduced personality will speak and act in response to future events.

[1262] The "means for asking questions taking into consideration the emotional state of the personality" refers to a means for recognizing the emotional state of the reproduced personality and asking questions in accordance with that state.

[1263] "Means for generating reactions of the personality using a generative AI model" refers to means for generating reactions of the reproduced personality using a generative AI model.

[1264] The "means for generating a question for the character using a prompt sentence" refers to a means for generating a question for the reproduced character using a prompt sentence.

[1265] The "means for determining the emotion based on the reaction of the personality" is a means for analyzing the reaction of the reproduced personality and determining its emotion.

[1266] The following system configuration will be described as an embodiment of the present invention.

[1267] System Configuration

[1268] The system consists of a server and a user terminal. The server has the function of learning the records of a specific person and regenerating a thought clone based on those records. It also has the function of asking questions to the regenerated personality, generating responses, and determining emotions. The user terminal provides an interface for the user to input questions and receive responses from the regenerated personality.

[1269] Hardware and software used

[1270] Hardware: Servers, smartphones

[1271] Software: Python, OpenAI API

[1272] Data processing and calculation

[1273] 1. A means of studying the records of specific individuals

[1274] The server collects information about a specific person and uses that information to learn about that person's intelligence, thoughts, feelings, emotions, morals, and ethics, thereby creating a detailed profile of that person.

[1275] 2. A means of recreating a thought clone of the individual based on the learned records

[1276] The server uses the records of a specific person it has learned to create a clone that mimics that person's thoughts and actions. This clone understands the specific person's past behavioral principles and, by providing them with current information, simulates what they would say and do in response to future events.

[1277] 3. A means of questioning the regenerated personality

[1278] A question entered from the user's device is sent to the server, which then asks the question to the reproduced personality. The question is generated as a prompt sentence, and a generative AI model is used to generate a response from the reproduced personality.

[1279] 4. A method for asking questions that take into account the emotional state of the individual in question

[1280] The server recognizes the emotional state of the recreated personality and generates questions accordingly. For example, if the recreated personality expresses joy, the server will ask a question such as, "What made you happy?"

[1281] 5. Means for generating the personality's responses using a generative AI model

[1282] The server uses a generative AI model to generate reactions for the reproduced personality, which are then sent to the user's device and displayed to the user.

[1283] 6. Means for generating questions for the personality using prompt sentences

[1284] The server uses prompts to generate questions for the persona, such as "What do you think about smartphones, which didn't exist in your lifetime?"

[1285] 7. Means for determining emotions based on the personality's reactions

[1286] The server analyzes the reactions of the reproduced personality and determines its emotions. For example, if the reactions contain keywords such as "happy" or "sad," the server determines the emotions based on those keywords.

[1287] Specific examples

[1288] Consider a case where a user uses a smartphone to ask a question to a specific historical figure. The user inputs the question, "What do you think about smartphones, which didn't exist in your time?" The server generates this question as a prompt sentence and uses a generative AI model to generate a response from the reproduced personality. The generated response might be something like, "Modern technology is very interesting. It would have been unthinkable in my time." The server then analyzes this response and determines its emotion. For example, if the response indicates a positive emotion such as "interesting," the server determines it as "joy."

[1289] Example prompt sentence:

[1290] What do you think about smartphones, which didn't exist in your lifetime?

[1291] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1292] Step 1:

[1293] The user uses the terminal to enter a question.

[1294] Input: A question typed in by the user (e.g., "What do you think about smartphones, which didn't exist in your lifetime?")

[1295] Output: Question text

[1296] Specific actions: The user opens the application on their smartphone, enters a question in the text box for entering a question, and presses the send button.

[1297] Step 2:

[1298] The terminal sends the entered question to the server.

[1299] Input: Question text

[1300] Output: Question data sent to the server

[1301] Specific operation: The terminal sends the entered question text to the server as an HTTP request.

[1302] Step 3:

[1303] The server receives the question and generates a prompt.

[1304] Input: Question data sent to the server

[1305] Output: Prompt statement (e.g., "What do you think about smartphones, which didn't exist in your lifetime?")

[1306] Specific behavior: The server parses the received question data and formats it as a prompt.

[1307] Step 4:

[1308] The server uses a generative AI model to generate the reactions of the played personality.

[1309] Input: prompt statement

[1310] Output: Response of recreated personality (e.g., "Modern technology is very interesting. It was unthinkable in my time.")

[1311] Specific operation: The server calls the OpenAI API, passes the prompt sentence as input to the generative AI model, and receives the generated response.

[1312] Step 5:

[1313] The server analyzes the reactions of the recreated personality and determines its emotions.

[1314] Input: Response from recreated personality

[1315] Output: Emotion determination result (e.g., "joy")

[1316] What happens: The server analyzes the reaction text and determines the sentiment based on specific keywords (e.g., "interesting").

[1317] Step 6:

[1318] The server transmits the generated reaction and emotion determination result to the terminal.

[1319] Input: Generated reaction, emotion judgment result

[1320] Output: Reaction and emotion data sent to the device

[1321] Specific operation: The server sends the generated reaction and emotion judgment result to the terminal as an HTTP response.

[1322] Step 7:

[1323] The terminal displays the received response and emotion determination result to the user.

[1324] Input: Reaction data and emotion data sent to the device

[1325] Output: Reactions displayed to the user and emotion judgment results

[1326] Specific operation: The device analyzes the received data and displays the response text and emotion determination result on the user interface.

[1327] Example 3

[1328] Next, a description will be given of Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1329] In conventional systems, when recreating the thoughts and emotions of a specific person, it was difficult to take into account the latest information and the person's emotional state. As a result, the recreated personality could not respond appropriately to real-world social situations or technological advances, making it impractical. In addition, because information was not provided according to the person's emotional state, the recreated personality's reactions sometimes lacked consistency.

[1330] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1331] In this invention, the server includes means for learning the records of a specific person, means for recreating a thought clone of the person based on the learned records, means for asking questions to the recreated personality, means for understanding and verifying the personality's past behavioral principles, means for providing the personality with current information, means for eliciting how the personality will speak / act in response to potential future events, means for collecting the latest information, means for analyzing the collected information and extracting important keywords and topics, means for generating prompts based on the extracted keywords and topics, means for inputting the generated prompts and predicting how the specific personality will react to the information, and means for recognizing emotional states and providing information corresponding to the recognized emotional states. This allows the recreated personality to respond appropriately to real social situations and technological advances, and to provide consistent information according to emotional states.

[1332] A "record of a specific person" is a collection of information about a specific person, including their intelligence, thoughts, feelings, emotions, morals, and ethics.

[1333] "Means of learning" are methods and techniques for analyzing the records of a particular person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[1334] A "thought clone" is an imitation of a specific person's thoughts and emotions, recreated based on their records.

[1335] "Means of asking questions" are methods or techniques for asking specific questions to the recreated personality and eliciting its response.

[1336] "Past behavioral principles" are basic principles or patterns that indicate how a particular person has thought or acted in the past.

[1337] "Current information" refers to information about current social conditions and technological advances, such as the latest news articles and research papers.

[1338] "Means of collecting the latest information" refers to methods and techniques for collecting the latest news articles and research papers using web scraping tools and APIs.

[1339] "Means of analyzing information" refers to methods and techniques for analyzing collected information using natural language processing tools and extracting important keywords and topics.

[1340] "Means for generating prompt sentences" refers to methods or techniques for creating prompt sentences to be input into generative AI models based on extracted keywords and topics.

[1341] "Means for predicting responses" refers to methods or techniques that input the generated prompt sentences into an AI model and predict how a specific personality will respond to that information.

[1342] "Means for recognizing emotional states" refers to methods or techniques for recognizing the emotional states of the reproduced personalities using emotion analysis tools.

[1343] "Means of providing information" refers to methods or techniques for selecting and providing appropriate information to the reproduced personality according to the recognized emotional state.

[1344] This invention is a system that learns the records of a specific person and reproduces that person's thought clone. The system consists of three main elements: a server, a terminal, and a user.

[1345] The server first collects and studies the records of a specific person. This includes information about the person's intelligence, thoughts, emotions, feelings, morals, and ethics. The server analyzes this information and generates data to recreate a thought clone of the specific person.

[1346] The server then collects information such as the latest news articles and research papers using a web scraping tool or an API (e.g., NewsAPI) and stores the collected information in a database (e.g., MySQL).

[1347] The server analyzes the collected information using natural language processing (NLP) tools to extract important keywords and topics, and generates prompts based on the analysis results to be input into a generative AI model (e.g., GPT-4).

[1348] The device inputs the generated prompt sentences into the AI ​​model to predict how a specific personality will react to the information, and the prediction results are provided to the user through the device's user interface.

[1349] Furthermore, the server uses emotion analysis tools (e.g., IBM Watson Tone Analyzer or Microsoft Azure) to recognize the emotional state of the person. Depending on the emotional state, the server selects appropriate information and provides it to the user through the terminal.

[1350] As a concrete example, consider the case where the latest news item is "A vaccine for the new coronavirus has been developed." The server analyzes this information and generates the following prompt:

[1351] "How do you feel when you hear the news that a vaccine for COVID-19 has been developed?"

[1352] The device then inputs this prompt into an AI model to predict how a particular personality will respond—for example, the AI ​​model might predict "feel joy."

[1353] Furthermore, using emotion analysis tools, if the personality is actually expressing joy, the server will provide information that amplifies the joy, such as "The successful development of a COVID-19 vaccine has increased the chances of saving people around the world."

[1354] In this way, the system can provide appropriate information according to the emotional state of the personality. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1355] Step 1:

[1356] The server collects and learns records of specific people.

[1357] Input: Information containing a specific person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[1358] Specific operation: The server retrieves data about a specific person from the database and analyzes it using a machine learning algorithm, thereby generating data for recreating a thought clone of that specific person.

[1359] Output: Data to recreate a thought clone of a specific person.

[1360] Step 2:

[1361] The server collects information such as the latest news articles and research papers.

[1362] Input: URL of news site or academic paper site.

[1363] Specific operation: The server uses a web scraping tool or API (e.g., NewsAPI) to collect the latest information from the specified site. The collected information is stored in a database.

[1364] Output: Collected news articles and research paper data.

[1365] Step 3:

[1366] The server analyzes the collected information and extracts important keywords and topics.

[1367] Input: Data collected from news articles and research papers.

[1368] How it works: The server uses natural language processing (NLP) tools to analyze the collected information, extracting frequently occurring keywords and topics.

[1369] Output: Extracted keywords and topics.

[1370] Step 4:

[1371] The server generates a prompt sentence based on the extracted keywords and topics.

[1372] Input: Extracted keywords or topics.

[1373] Specific operation: The server creates a prompt to be input to a generative AI model (e.g., GPT-4). For example, in response to the information, "A vaccine for the new coronavirus has been developed," the server generates the prompt, "How do you feel when you hear the news that a vaccine for the new coronavirus has been developed?"

[1374] Output: The generated prompt statement.

[1375] Step 5:

[1376] The device inputs the generated prompts into an AI model that predicts how a particular personality will react to the information.

[1377] Input: The generated prompt statement.

[1378] Specific operation: The device inputs the prompt sentence received from the server into the AI ​​model, which then predicts the response of a specific personality based on the input prompt sentence.

[1379] Output: Predicted response.

[1380] Step 6:

[1381] The server uses an emotion analysis tool to recognize the emotional state of the persona.

[1382] Input: predicted response.

[1383] Specific operation: The server uses an emotion analysis tool (e.g., IBM Watson Tone Analyzer or Microsoft Azure) to analyze the predicted reaction and recognize the emotional state.

[1384] Output: Perceived emotional state.

[1385] Step 7:

[1386] The server selects appropriate information according to the recognized emotional state and provides it to the user through the terminal.

[1387] Input: Perceived emotional state.

[1388] Specific operation: The server searches the database for and selects information that corresponds to the emotional state. For example, if the personality expresses joy, the server selects information that amplifies joy, such as "The successful development of a COVID-19 vaccine has increased the chances of saving people around the world." The selected information is sent to the device and displayed to the user.

[1389] Output: Pertinent information provided to the user.

[1390] (Application example 3)

[1391] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1392] Conventional systems were capable of reproducing thought clones based on the records of specific individuals, and understanding and verifying past behavioral principles, but lacked the ability to recognize the user's emotional state in real time and provide information accordingly. This meant that they were unable to provide appropriate information based on the user's emotions, resulting in a poor user experience. Furthermore, the lack of information provided based on the user's emotional state made it difficult to provide services that were sensitive to the user's emotions.

[1393] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1394] In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of the person based on the learned records, means for asking questions to the reproduced personality, means for understanding and verifying the personality's past principles of behavior, means for providing current information to the personality, means for eliciting how the personality will speak / act in response to future events, means for recognizing the user's emotional state, and means for providing information based on the recognized emotional state, thereby making it possible to provide appropriate information according to the user's emotional state.

[1395] A "means for learning records of a specific person" is a device or program that collects information about a specific person's intelligence, thoughts, feelings, emotions, morals, ethics, etc., and learns based on this information.

[1396] "Means for reproducing a thought clone of a person based on learned records" refers to a device or program that generates a clone that imitates the thoughts and actions of a specific person based on learned records of that person.

[1397] "Means for asking questions to the regenerated personality" is a device or program for asking questions to the regenerated thought clone and obtaining its response.

[1398] "Means for understanding and verifying the personality's past behavioral principles" refers to a device or program that analyzes the past behavior and thought patterns of the reproduced thought clone, and understands and verifies those principles.

[1399] A "means for providing current information to said personality" is a device or program for providing current information to the recreated thought clone, such as the latest news articles or research papers.

[1400] "Means to derive how the personality will speak / act in response to future events" refers to a device or program that predicts how the regenerated thought clone will speak and act in response to future events and derives the results.

[1401] "Means for recognizing the user's emotional state" refers to a device or program that analyzes the user's facial expressions, voice, actions, etc., and recognizes their emotional state in real time.

[1402] A "means for providing information based on a recognized emotional state" is a device or program for providing appropriate news articles or content in response to a recognized emotional state of a user.

[1403] The system for implementing this invention has the ability to learn the records of a specific person and reproduce thought clones based on those records, as well as recognize the user's emotional state in real time and provide information according to that emotion.

[1404] Hardware and software used

[1405] Hardware: Smartphones, servers

[1406] Software: Python, EmotionRecognizer, NewsAPI

[1407] System configuration

[1408] 1. Means of studying a particular person's record:

[1409] The server collects information on a specific person's intelligence, thoughts, emotions, feelings, morals, and ethics, and uses this information to learn, thereby understanding the behavior and thought patterns of that specific person.

[1410] 2. A means of recreating a thought clone of the individual based on the learned records:

[1411] The server generates a clone that mimics the thoughts and actions of a specific person based on the learned records of that person. This clone can reproduce the specific person's past behavioral principles.

[1412] 3. Means of questioning the regenerated personality:

[1413] The server asks questions to the regenerated thought clone and receives its responses. These questions are used to understand and verify the past behavioral principles of a specific person.

[1414] 4. Means of understanding and verifying the individual's past motivations:

[1415] The server analyzes the past actions and thought patterns of the recreated thought clone, and understands and verifies its principles, which allows for a deep understanding of the principles of a specific person's behavior.

[1416] 5. Means of providing current information to the individual:

[1417] The server provides the regenerated thought clone with current information, such as the latest news articles and research papers, allowing it to predict how the thought clone will react to current social conditions and advances in science and technology.

[1418] 6. Methods to elicit how the individual will respond to future events:

[1419] The server predicts how the regenerated thought clone will speak and act in response to future events and derives the results.

[1420] 7. Means of recognizing the user's emotional state:

[1421] The smartphone analyzes the user's facial expressions, voice, and behavior to recognize their emotional state in real time. It uses EmotionRecognizer to accurately grasp the user's emotions.

[1422] 8. Means of providing information based on perceived emotional states:

[1423] The server provides appropriate news articles and content according to the user's recognized emotional state. It analyzes the emotions of news articles and filters information that matches the user's emotions.

[1424] Specific examples

[1425] Emotion recognition is performed through the camera and microphone while the user is using the smartphone.

[1426] If the user is feeling "joyed," positive news articles (e.g., "New technology is changing lives") will be displayed.

[1427] If the user is feeling "sad," they will see comforting and encouraging news articles (e.g., "Community overcomes difficulties").

[1428] Prompt Sentence Examples

[1429] "Generate positive news articles to show users when they're feeling joy."

[1430] "Generate comforting news articles to show users when they're feeling sad."

[1431] In this way, a personalized news feed can be provided according to the user's emotional state.

[1432] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1433] Step 1:

[1434] The server collects records of specific individuals. As input, it receives information about the individual's intelligence, thoughts, emotions, feelings, morals, and ethics. Based on this data, it learns the individual's behavior and thought patterns. As output, it generates a record of the learned individual.

[1435] Step 2:

[1436] The server reproduces a thought clone of a specific person based on the learned record. As input, it receives the learned record generated in step 1. Based on this, it generates a clone that imitates the thoughts and actions of the specific person. As output, it obtains the reproduced thought clone.

[1437] Step 3:

[1438] The server asks questions to the reproduced thought clone. As input, it receives the reproduced thought clone and the question. This gives a response to understand and verify the past behavioral principles of a specific person. As output, the response of the thought clone is generated.

[1439] Step 4:

[1440] The server understands and verifies the past behavioral principles of the reproduced thought clone. As input, it receives the response obtained in step 3. It analyzes this and processes the data to understand the behavioral principles of a specific person. As output, it obtains the understanding results of the behavioral principles.

[1441] Step 5:

[1442] The server provides current information, such as the latest news articles and research papers, to the regenerated thought clone. As input, it receives the current information and the regenerated thought clone. This predicts how the thought clone will react to current social conditions and advances in science and technology. As output, it obtains the thought clone's reaction.

[1443] Step 6:

[1444] The server predicts how the regenerated thought clone will speak and act in response to future events, and derives the results. As input, it receives the reaction of the thought clone obtained in step 5. Based on this, it predicts what the thought clone will say and do in response to future events. The predicted result is obtained as output.

[1445] Step 7:

[1446] The device recognizes the user's emotional state in real time. As input, it receives data such as the user's facial expressions, voice, and behavior. It uses EmotionRecognizer to analyze this data and recognize the user's emotional state. As output, it obtains the recognized emotional state.

[1447] Step 8:

[1448] The server provides appropriate news articles and content according to the recognized emotional state of the user. As input, it receives the emotional state obtained in step 7 and news article data. It analyzes the emotions of the news articles and filters information that matches the user's emotions. As output, it obtains a personalized news feed.

[1449] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1450] The data generation model 58 is a so-called generative AI (Artificial Intelligence).

[1451] An example of a data generation model58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ) and other generative AIs. Data generation models 58

[1452] , obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in the data format of voice data, text data, etc. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1453] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1454] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1455] [Third embodiment]

[1456] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1457] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1458] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1459] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1460] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1461] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1462] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1463] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1464] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1465] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1466] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1467] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1468] "Example 1"

[1469] In one embodiment of the present invention, a machine learning model is developed to learn the records of a specific individual by collecting digital data such as the individual's past actions, statements, books, papers, and social media posts. This data is then input into a machine learning model that learns the individual's thought patterns and principles of behavior, enabling the reproduction of that individual's personality.

[1470] "Example 2"

[1471] Natural language processing technology is used to provide an interface that allows people to naturally ask questions to the recreated personality. For example, by asking a question such as, "What do you think about smartphones, which didn't exist in your lifetime?", it is possible to predict how the person will react to modern technology.

[1472] "Example 3"

[1473] To provide current information to the personality, information such as the latest news articles and research papers is input into the system, which can then predict how the person will react to current social conditions and advances in science and technology.

[1474] The processing flow of each embodiment will be described below.

[1475] "Example 1"

[1476] Step 1: Collect digital data such as a specific person's past actions, statements, works such as books and papers, and social media posts.

[1477] Step 2: Preprocess the collected data and convert it into a format suitable for the machine learning model.

[1478] Step 3: The preprocessed data is input into the machine learning model and the model is trained. As a result of the training, a model that has learned a person's thought patterns and principles of behavior is obtained.

[1479] "Example 2"

[1480] Step 1: Using natural language processing technology, develop an interface that allows humans to ask questions naturally.

[1481] Step 2: Ask the recreated personality questions through the developed interface. Step 3: Analyze the responses from the recreated personality and predict how the person will react to modern technology.

[1482] "Example 3"

[1483] Step 1: Gather information such as the latest news articles and research papers.

[1484] Step 2: Enter the collected information into the system.

[1485] Step 3: Based on the input information, predict how the recreated personality will react to current social conditions and advances in science and technology.

[1486] Example 1

[1487] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1488] Conventional systems have had difficulty accurately reproducing a specific person's thought patterns and principles of behavior. Furthermore, preprocessing of collected data and building, training, and evaluating machine learning models were often done manually, resulting in inefficiencies. Furthermore, the process of posing appropriate questions to the reproduced personality and generating responses was complicated.

[1489] The identification process by the identification processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means. In this invention, the server includes a means for collecting digital data and performing preprocessing, a means for constructing and training a machine learning model, a means for evaluating the trained model, and a means for inputting a prompt sentence into the generative AI model and generating a response. This makes it possible to efficiently and accurately reproduce the thought patterns and principles of behavior of a specific person, ask appropriate questions to the reproduced personality, and generate a response.

[1490] A "specific person" refers to a specific individual, and is the subject of collection of digital data such as that individual's past actions, statements, copyrighted works, and social media posts.

[1491] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works, and social media posts.

[1492] "Thought clone" refers to a machine learning model built to replicate a specific person's thought patterns and behavioral principles.

[1493] "Digital data" refers to electronically stored information such as a specific person's past actions, statements, copyrighted works, and social media posts.

[1494] "Preprocessing" refers to the process of cleaning, tokenizing, normalizing, and otherwise converting collected digital data into a format suitable for machine learning models.

[1495] A "machine learning model" refers to an algorithm or neural network used to learn and replicate a specific person's thought patterns and behavioral principles.

[1496] "Learning" refers to the process of inputting data into a machine learning model so that the model can understand and reproduce a specific person's thought patterns and principles of behavior.

[1497] "Evaluation" refers to the testing and verification process to confirm the predictive accuracy and performance of a trained machine learning model.

[1498] "Generative AI model" refers to an artificial intelligence model that is used to reproduce a specific person's thought patterns and principles of behavior and generate responses to prompts.

[1499] A "prompt" refers to a question or instruction entered into a generative AI model, and is input data used to generate a response based on a specific person's thought patterns.

[1500] This invention is a system that learns the records of a specific person and reproduces that person's thought patterns and principles of behavior. A specific embodiment of this system will be described below.

[1501] First, the user collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. This data serves as the basis for learning the person's thought patterns and principles of behavior. The collected data is then sent to a server.

[1502] The server then preprocesses the collected digital data, specifically cleaning, tokenizing, and normalizing the text data. The software used includes Python libraries, such as removing unnecessary HTML tags from social media posts and splitting the text into words and sentences.

[1503] Using the preprocessed data, the server builds a machine learning model. This model is used to learn a specific person's thought patterns and behavioral principles. Specifically, it uses a deep learning framework such as TensorFlow. Building the model includes dividing the dataset, selecting a model, and designing a neural network.

[1504] The server then inputs the preprocessed data into the machine learning model it has built and trains the model. The model is trained using a training set and an optimization algorithm such as gradient descent is used to minimize the loss function.

[1505] Once the training is complete, the server evaluates the model. The evaluation checks the model's predictive accuracy using a test set. Evaluation metrics include precision, recall, and F1 score.

[1506] Finally, the user uses the trained model to recreate a specific person's personality. Specifically, the user inputs a prompt into the generative AI model, which simulates how the specific person would respond. For example, the user can input the following prompt:

[1507] "Tell me about your latest book."

[1508] The generative AI model recreates the thought patterns of a specific person based on the input prompt and generates a response that is highly realistic because it is based on the specific person's past behavioral principles and thought patterns.

[1509] The above is a specific embodiment for carrying out the present invention. This system makes it possible to efficiently and accurately reproduce the thought patterns and behavioral principles of a specific person, pose appropriate questions to the reproduced personality, and generate responses.

[1510] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1511] Step 1: Data collection

[1512] A user collects digital data such as a specific person's past actions, statements, copyrighted works, and social media posts. The input is digital data collected from the internet or databases. The output is the collected digital data sent to a server. Specifically, a web scraping tool is used to automatically collect social media posts.

[1513] Step 2: Data Preprocessing

[1514] The server preprocesses the collected digital data. The input is the collected digital data. The output is the preprocessed data. Specific operations include cleaning, tokenizing, and normalizing the text data. For example, a Python library is used to tokenize the text and remove unnecessary HTML tags.

[1515] Step 3: Building a machine learning model

[1516] The server builds a machine learning model using the preprocessed data. The input is the preprocessed data. The output is the built machine learning model. Specifically, it designs a neural network model using a deep learning framework such as TensorFlow. It splits the dataset into a training set and a test set, and selects and designs a model.

[1517] Step 4: Training the model

[1518] The server inputs the preprocessed data into the constructed machine learning model and trains the model. The inputs are the preprocessed data and the constructed model. The output is a trained model. Specifically, the model is trained using a training set, and an optimization algorithm such as gradient descent is used to minimize the loss function.

[1519] Step 5: Evaluate the model

[1520] The server evaluates the model after training. The inputs are the trained model and a test set. The output is the model's predictive accuracy and performance. Specifically, it checks the model's predictive accuracy using the test set and calculates evaluation metrics such as precision, recall, and F1 score.

[1521] Step 6: Personality Regeneration

[1522] The user uses a trained model to recreate the personality of a specific person. The inputs are the trained model and a prompt. The output is a generated response. Specifically, the prompt is input into the generative AI model, and a simulation is performed to see how the specific person would respond. For example, the prompt is input as "Tell me about your latest book," and the model generates a response.

[1523] (Application example 1)

[1524] Next, a description will be given of Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1525] In today's world, there is a demand for systems that can reproduce the knowledge and thought patterns of specific famous people or historical figures, allowing users to converse with them. However, with conventional technology, it is difficult to effectively collect digital data about a specific person, input it into a machine learning model, and generate answers to user questions that are appropriate for that person. Therefore, the challenge is to provide a system that can accurately reproduce the thought patterns and principles of behavior of a specific person, allowing users to converse with that person.

[1526] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1527] In this invention, the server includes a means for learning the records of a specific person, a means for reproducing a thought clone of the person based on the learned records, and a means for asking questions to the reproduced personality. This makes it possible to collect digital data of a specific person, input it into a machine learning model for learning, and generate answers to user questions that are characteristic of that person.

[1528] "Specific person" refers to a specific individual, such as a particular celebrity or historical figure.

[1529] "Records" refers to digital data such as a specific person's past actions, statements, copyrighted works, and social media posts.

[1530] A "thinking clone" refers to an artificial intelligence model that reproduces the thought patterns and behavioral principles of a specific person.

[1531] A "machine learning model" refers to an algorithm or system that learns a specific person's thought patterns and behavioral principles based on collected digital data.

[1532] "Digital data" refers to electronically stored information such as a specific person's past actions, statements, copyrighted works, and social media posts.

[1533] "User" means an individual who uses the System to interact with a specific person.

[1534] "Questions" refer to questions that the user poses to the recreated personality.

[1535] "Answer" refers to the response that the recreated personality generates to the user's question.

[1536] A "thinking pattern" refers to how a particular person tends to think and judge a particular situation or problem.

[1537] A "principle of action" refers to a basic principle of how a particular person should act in a particular situation or problem.

[1538] To implement this invention, it is necessary to build a system that collects digital data of specific individuals and inputs it into a machine learning model for learning. This system consists of three main elements: a server, a terminal, and a user.

[1539] server

[1540] The server includes a means for learning the records of a specific person, a means for reproducing a thought clone of that person based on the learned records, and a means for asking questions to the reproduced personality. Specifically, the server performs the following processes.

[1541] 1. Data collection: The server collects digital data such as social media posts, copyrighted works, and statements of specific individuals. This data is collected automatically using APIs.

[1542] 2. Data input: Convert the collected data into JSON format and input it into a machine learning model such as OpenAI's GPT-3.

[1543] 3. Model learning: The GPT-3 model learns a specific person's thought patterns and behavioral principles based on the input data.

[1544] 4. Response generation: The trained model is used to generate a response that is appropriate for the user's question.

[1545] Terminal

[1546] A terminal provides an interface for users to interact with the system. Devices such as smartphones and tablets fall into this category. Terminals have the following functions:

[1547] 1. User Interface: Provides an interface for users to enter questions and receive answers.

[1548] 2. Data transmission: The question entered by the user is sent to the server.

[1549] 3. Data reception: Receives the response sent from the server and displays it to the user.

[1550] User

[1551] A user is an individual who uses the system to interact with a specific person. The user performs the following operations:

[1552] 1. Question input: The user inputs a question for a specific person through the terminal interface.

[1553] 2. Receiving the answer: The user receives the answer sent from the server on their terminal and checks it.

[1554] Specific examples

[1555] For example, if a user selects "famous scientists" as a specific person, the user can enter a question such as:

[1556] "Tell me about your next project."

[1557] "Tell me about the theory of relativity."

[1558] The server uses the trained model to generate answers to these questions and provides them to the user via the terminal.

[1559] Prompt Sentence Examples

[1560] "I discovered something new today."

[1561] I'm thinking about my next project.

[1562] "Science is constantly evolving."

[1563] In this way, the user can experience the knowledge and thought patterns of a particular person and enjoy the dialogue.

[1564] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1565] Step 1:

[1566] The server collects digital data of a specific person. Specifically, it automatically collects data such as social media posts, copyrighted works, and statements using an API. The input is the person's identifying information, and the output is the collected digital data.

[1567] Step 2:

[1568] The server converts the collected digital data into JSON format. The input is the collected digital data, and the output is JSON format data. This conversion makes the data suitable for machine learning models.

[1569] Step 3:

[1570] The server inputs the JSON-formatted digital data into a machine learning model such as OpenAI's GPT-3, training the model. The input is JSON-formatted data, and the output is a trained model. The model learns a specific person's thought patterns and principles of behavior.

[1571] Step 4:

[1572] A user inputs a question for a specific person through the terminal interface. The input is the user's question, and the output is the question data sent from the terminal to the server.

[1573] Step 5:

[1574] The terminal sends the question entered by the user to the server. The input is the user's question data, and the output is the data to be sent to the server.

[1575] Step 6:

[1576] The server generates answers to user questions using the trained model. The input is the user's question data and the trained model, and the output is the generated answer. The server inputs the question as a prompt to the model and generates an answer.

[1577] Step 7:

[1578] The server sends the generated answer to the terminal. The input is the generated answer, and the output is the data sent to the terminal.

[1579] Step 8:

[1580] The terminal receives the response sent from the server and displays it to the user. The input is the data sent from the server, and the output is the response displayed to the user.

[1581] In this way, the user can experience the knowledge and thought patterns of a particular person and enjoy the dialogue.

[1582] Example 2

[1583] Next, a description will be given of Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1584] In conventional systems, the process of asking questions to the reproduced personality and receiving responses was complicated, making it difficult for users to ask questions naturally. Also, it was difficult to predict how the reproduced personality would respond to modern technology and information.

[1585] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for a user to input a question, a means for a terminal to send a question to the server, a means for the server to analyze the question, a means for the server to generate a prompt sentence for the generative AI model, a means for the server to input a prompt sentence to the generative AI model, a means for the server to receive a generated response, a means for the server to send the response to the terminal, and a means for the terminal to display the response to the user. This allows the user to naturally ask a question and easily obtain a response from the reproduced personality to the question.

[1586] "Records of a specific person" are data that include information about a specific person's intelligence, thoughts, feelings, emotions, morals, ethics, etc.

[1587] "Means of learning" are methods and techniques for analyzing the records of a particular person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[1588] "Means for reproducing thought clones" refers to methods and techniques for generating artificial intelligence that imitates the thoughts and actions of a specific person based on the records of that person that it has learned.

[1589] A "means for asking a question" is a method or technique that allows a user to naturally input a question to the reproduced personality and transmit the question to the system.

[1590] "Means for understanding and verifying the principles of past behavior" refers to methods and techniques for analyzing the past behavior and thought patterns of the recreated personality and predicting responses to current questions based on that analysis.

[1591] "Current means of providing information" are methods and techniques for providing modern technology and information to the recreated personality and eliciting its response.

[1592] "Means of eliciting how to speak / act in response to possible future events" refers to methods or techniques for predicting how the recreated personality will react to future events and generating that response.

[1593] The "means for the user to input a question" is an interface that allows the user to input a question to the reproduced personality using a terminal.

[1594] The "means by which the terminal transmits the question to the server" refers to the communication protocol or technology used to transmit the question entered by the user from the terminal to the server.

[1595] The "means by which the server analyzes the question" refers to a method or technology by which the server analyzes the question received by using natural language processing technology and extracts the grammatical structure and keywords of the question.

[1596] "Means for generating prompt sentences for a generative AI model" refers to methods or technologies for generating appropriate prompt sentences for a generative AI model based on the analysis results.

[1597] "Means for inputting prompt sentences into a generative AI model" refers to methods or techniques for inputting generated prompt sentences into a generative AI model and obtaining a response from the model.

[1598] "Means for receiving the generated response" refers to the method or technology by which the server receives the response from the generative AI model.

[1599] The "means for sending the response to the terminal" refers to the communications protocol or technology by which the server sends the generated response to the user's terminal.

[1600] The "means by which the terminal displays the response to the user" is an interface for displaying the response received by the terminal to the user.

[1601] This invention is a system for posing questions to a recreated personality and obtaining responses. The system includes means for learning a record of a specific person and recreating a thought clone based on the record. The system also includes means for posing questions to the recreated personality and generating responses to the questions.

[1602] Hardware and software used

[1603] Hardware: Servers, user devices (PCs, smartphones, etc.)

[1604] Software: Natural language processing libraries, generative AI models (e.g., GPT-4)

[1605] Specific operation of the system

[1606] 1. User enters a question:

[1607] The user uses the device to input questions for the recreated personality, such as "What do you think about smartphones, which didn't exist in your lifetime?"

[1608] 2. The device sends a question to the server:

[1609] The terminal sends the question entered by the user to the server using a communication protocol such as an HTTP request.

[1610] 3. The server parses the question:

[1611] The server analyzes the received question using a natural language processing library. Specifically, it analyzes the grammatical structure of the question and extracts keywords. For example, keywords such as "smartphone" and "What do you think?" are extracted.

[1612] 4. The server generates a prompt for the generative AI model:

[1613] Based on the analysis results, the server generates a prompt for the generative AI model (e.g., GPT-4). For example, it generates a prompt such as, "You are a person from the 19th century. What do you think about modern smartphones?"

[1614] 5. The server inputs a prompt to the generative AI model:

[1615] The server inputs the generated prompt sentence into the generative AI model, which then generates a response based on the prompt sentence.

[1616] 6. The server receives the generated response:

[1617] The server receives a response from the generative AI model, such as, "I think smartphones are very useful tools, but they were unimaginable in my day."

[1618] 7. The server sends the response to the device:

[1619] The server then sends the generated response to the user's device, again using a communication protocol such as an HTTP response.

[1620] 8. The terminal displays the response to the user:

[1621] The terminal displays the received response to the user, who can then check the response from the personality played back on the screen.

[1622] Specific examples

[1623] User question: "What do you think about smartphones, which didn't exist in your lifetime?"

[1624] Prompt for generative AI model: "You are a person from the 19th century. What do you think about modern smartphones?"

[1625] This system allows users to naturally ask questions to the reproduced personality and easily obtain responses to those questions.

[1626] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1627] Step 1:

[1628] The user enters a question.

[1629] The user inputs a question using the device's keyboard or touch screen. Once input is complete, the user clicks the "Submit" button. The input data is a text-based question. For example, a question might be entered such as, "What do you think about smartphones, which didn't exist when you were alive?"

[1630] Step 2:

[1631] The terminal sends a question to the server.

[1632] The device converts the input question into JSON format and sends it to the server as an HTTP POST request. The input data is the question in text format, and the output data is the question data in JSON format.

[1633] Step 3:

[1634] The server parses the question.

[1635] The server parses the received JSON data and extracts the question. It uses a natural language processing library to analyze the grammatical structure of the question and extract keywords. The input data is the question data in JSON format, and the output data is the analysis result (grammatical structure and keywords). For example, keywords such as "smartphone" and "What do you think?" are extracted.

[1636] Step 4:

[1637] The server generates a prompt for the generative AI model.

[1638] Based on the analysis results, the server generates a prompt for the generative AI model (e.g., GPT-4). The input data is the analysis results, and the output data is the prompt. For example, it generates a prompt such as, "You are a person from the 19th century. What do you think about modern smartphones?"

[1639] Step 5:

[1640] The server inputs a prompt sentence into the generative AI model.

[1641] The server inputs the generated prompt sentence into the generative AI model and waits for the model's response. The input data is the prompt sentence, and the output data is the response from the generative AI model.

[1642] Step 6:

[1643] The server receives the generated response.

[1644] The server receives the response from the generative AI model and checks the response. The input data is the response from the generative AI model, and the output data is the checked response. For example, a response such as "I think smartphones are very useful tools, but they were unimaginable in my day" is generated.

[1645] Step 7:

[1646] The server sends a response to the terminal.

[1647] The server converts the generated response back into JSON format and sends it to the user's device as an HTTP response. The input data is the confirmed response, and the output data is the JSON formatted response data.

[1648] Step 8:

[1649] The terminal displays the response to the user.

[1650] The device parses the received JSON data and displays the response on the screen. The input data is the response data in JSON format, and the output data is the response in text format. The user can check the displayed response and enjoy interacting with the reproduced personality.

[1651] (Application example 2)

[1652] Next, a description will be given of Application Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1653] Previous systems had issues with the interactive experience, as it was difficult for users to ask questions naturally to the reproduced personality and there was no function to play back the answers aloud. Furthermore, the user experience was limited because there was no way for users to ask questions by voice or text input.

[1654] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for learning the records of a specific person, means for reproducing a thought clone of the person based on the learned records, means for posing questions to the reproduced personality, means for understanding and verifying the personality's past behavioral principles, means for providing the personality with current information, means for eliciting how the personality will speak / act in response to potential future events, means for reproducing the generated answers aloud using speech synthesis technology, and means for the user to pose questions by voice input or text input. This allows the user to naturally pose questions and reproduce the generated answers aloud, enabling a more interactive and richer experience.

[1655] "Records of a specific person" refers to information about a specific person, including their intelligence, thoughts, feelings, emotions, morals, and ethics.

[1656] "Means of learning" refers to techniques for analyzing records of a specific person and understanding that person's intelligence, thoughts, feelings, emotions, morals, and ethics.

[1657] A "thought clone" refers to an artificial intelligence that imitates the intelligence, thoughts, emotions, feelings, morals and ethics of a specific person based on learned records.

[1658] "Means for asking questions" refers to an interface that allows the user to naturally ask questions to the recreated personality.

[1659] "Past behavioral principles" refers to principles or patterns that show how a particular person has thought or acted in the past.

[1660] "Current information" refers to data about modern technology and events.

[1661] "Means of eliciting what to say / do" refers to technology that predicts how the recreated personality will react to future events and provides that reaction to the user.

[1662] "Speech synthesis technology" refers to the technology that converts text data into speech.

[1663] "Means for playing back by voice" refers to technology for providing the generated answer to the user by voice using speech synthesis technology.

[1664] "Voice or text input means for asking questions" refers to an interface that allows a user to ask questions to the played-out persona using voice or text.

[1665] The system for implementing this invention includes a program for learning the records of a specific person and recreating a thought clone based on the records. The system allows the user to naturally ask questions to the regenerated personality and provides a function for replaying the answers by voice.

[1666] Hardware and software used

[1667] Hardware: Smartphone

[1668] Software: Python, OpenAI API (GPT-3), speech synthesis library

[1669] Data processing and calculation

[1670] 1. User input: The user uses their smartphone to ask a question by voice or text input.

[1671] 2. Question processing: The question is sent to the server, and an answer is generated for the specified personality using OpenAI's GPT-3 model.

[1672] 3. Answer Generation: The GPT-3 model generates appropriate answers to questions based on records of a specific person.

[1673] 4. Speech synthesis: The generated answer is converted into speech using a library.

[1674] 5. Audio playback: The answer is played aloud to the user through the smartphone speaker.

[1675] Specific examples

[1676] A user opens the app and asks a question to a specific historical figure. For example, "What do you think about smartphones, which didn't exist in your time?" This question is sent to a server, and the GPT-3 model generates an answer for that person. The answer might be something like, "Smartphones allow instant access to information and will greatly contribute to scientific advancement." This answer is then played aloud using speech synthesis technology and provided to the user.

[1677] Prompt Sentence Examples

[1678] As a specific historical figure, answer the following question: What do you think about smartphones, which didn't exist in your lifetime?

[1679] In this way, users can naturally ask questions and have the generated answers played back to them aloud, resulting in a more interactive and enriching experience.

[1680] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1681] Step 1:

[1682] A user starts the application using a smartphone and selects a specific historical figure. Next, the user inputs a question by voice or text. This input data is text data containing the user's question.

[1683] Step 2:

[1684] The device sends the question entered by the user to the server. The server analyzes the received question and sends a prompt sentence to the generative AI model (GPT-3) to generate an answer as a specific historical figure. This prompt sentence is text data containing the user's question.

[1685] Step 3:

[1686] The server receives a response from the generative AI model (GPT-3), which generates an answer to the question based on the records of specific historical figures. This generated answer is returned to the server as text data.

[1687] Step 4:

[1688] The server uses a speech synthesis library (pyttsx3) to convert the received text data into speech data. The speech synthesis library analyzes the text data and generates the corresponding speech data.

[1689] Step 5:

[1690] The server transmits the generated voice data to the device, which then plays the received voice data and provides it to the user. The user can then listen to the audio of the specific historical figure's response through the smartphone speaker.

[1691] Step 6:

[1692] The user can then enter the question again or select a different historical figure, allowing the user to continue the interactive experience.

[1693] Example 3

[1694] Next, a third embodiment of the third embodiment will be described. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1695] While conventional systems can reproduce a specific person's past actions and thoughts, it is difficult to predict how that person will react to the latest information. Furthermore, there is a lack of concrete means for inputting information such as the latest news articles or research papers and predicting how a specific person will react to that information. This makes it difficult to accurately predict a specific person's future actions and statements.

[1696] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1697] In this invention, the server includes means for learning the records of a specific person, means for recreating a thought clone of the person based on the learned records, means for asking questions to the recreated personality, means for understanding and verifying the personality's past behavioral principles, means for providing current information to the personality, means for analyzing the information, means for generating prompt sentences based on the analysis results, means for inputting the generated prompt sentences into a generative AI model, means for obtaining prediction results from the generative AI model, and means for deriving how the personality will speak / act in response to possible future events. This makes it possible to predict how a specific person will react to the latest information.

[1698] A "specific person" is an individual person that the system targets, and is the subject of learning records of that person's past actions, thoughts, emotions, etc.

[1699] "Means of studying records" refers to techniques and methods for collecting information, including a specific person's intelligence, thoughts, feelings, emotions, morals, and ethics, and for analyzing and studying this information.

[1700] "Means for reproducing thought clones" refers to techniques or methods for reproducing the thought and behavior patterns of a specific person based on learned records.

[1701] "Means of asking questions" are techniques or methods for asking specific questions to the recreated personality and obtaining a response.

[1702] "Means for understanding and verifying behavioral principles" refers to techniques and methods for analyzing a specific person's past behavior and thought patterns, and understanding and verifying those principles.

[1703] "Current information delivery means" are techniques or methods for providing information, such as the latest news articles or research papers, to a particular person's recreated personality.

[1704] "Means of analyzing information" refers to techniques and methods for analyzing given information using natural language processing techniques, etc., and extracting important keywords and context.

[1705] "Means for generating prompt sentences" refers to the technology or method for creating prompt sentences to be input into a generative AI model based on the analysis results.

[1706] "Means for inputting into a generative AI model" refers to the technology or method for inputting the generated prompt sentence into a generative AI model and having it make a prediction.

[1707] "Means for obtaining the prediction results of a generative AI model" refers to the technology or method for obtaining the prediction results obtained from a generative AI model and using them within the system.

[1708] "Means to elicit speech / action" are techniques or methods for eliciting how a specific person's regenerated personality will speak or act in response to events that may occur in the future.

[1709] MODE FOR CARRYING OUT THE INVENTION

[1710] This invention is a system that learns the past actions and thoughts of a specific person and predicts how that person will react by providing the latest information to the reproduced personality. A specific embodiment of this system will be described below.

[1711] Hardware and software used

[1712] This system operates with three main parties: a server, a terminal, and a user. The server receives data, analyzes it, runs the generative AI model, and provides the results. The terminal provides an interface for users to input information and receive results. Users input information about specific people, the latest news articles, research papers, etc.

[1713] The server uses the following software and technologies:

[1714] Natural Language Processing (NLP) techniques: Using generative AI models such as Google's BERT and OpenAI's GPT-4.

[1715] Database: A database system for storing entered information.

[1716] Web Server: A web server for receiving input from users and providing results.

[1717] System Operation

[1718] A user uses a terminal to input the latest news articles or research papers into the system. For example, if a user inputs a news article about new developments in quantum computing, the following specific actions occur:

[1719] 1. A user copies and pastes a news article into the input form and clicks the submit button.

[1720] 2. The server receives the text data of the news article and stores it in a database.

[1721] 3. The server analyzes the news article using the BERT model and extracts important keywords and context.

[1722] 4. The server generates a prompt based on the analysis results, for example, "Predict how a particular scientist will react to new advances in quantum computing."

[1723] 5. The prompt sentence generated by the server is input into the generative AI model (GPT-4).

[1724] 6. The server obtains predictions from the generative AI model, such as "This scientist will be very excited about new advances in quantum computing and will believe they will have a major impact on future research."

[1725] 7. The server provides the prediction results to the user, who then checks the results through their device.

[1726] Specific examples

[1727] For example, if a user inputs "a news article about the latest quantum computing advances" into the system, the following prompt will be generated:

[1728] "Predict how a particular scientist will react to a new development in quantum computing."

[1729] The generative AI model (GPT-4) takes this prompt and predicts how a particular scientist will reac...

Claims

1. A first means for learning a record of a particular person, the record being data including information on intelligence, thoughts, feelings, emotions, morals, and ethics regarding the particular person; A second means for reproducing a thought clone of the specific person based on the learned record of the specific person; A third means for acquiring a user's question for the thought clone of the specific person; a fourth means for generating a first prompt sentence for outputting an answer to the question from the thought clone of the specific person based on the content of the question by the user; A fifth means for inputting the generated first prompt sentence into a generative AI model to generate an answer from a thought clone of the specific person; A sixth means for transmitting an answer from the generated thought clone of the specific person to the terminal of the user; A system including:

2. a seventh means for allowing the user to input current news articles and research papers; The fourth means generates a second prompt sentence based on the latest news article and the research paper, for outputting an opinion from the thought clone of the specific person regarding the latest news article and the research paper; The fifth means inputs the generated second prompt sentence into a generative AI model to generate an opinion from the thought clone of the specific person regarding the latest news articles and research papers. The system of claim 1 .

3. An eighth means for analyzing emotions of the specific person from past tweets and posts of the specific person using an emotion engine; A ninth means for making the thinking clone of the specific person respond according to the emotion of the specific person; 3. The system according to claim 1 or 2, comprising:

Images