system

Abstract

システムを提供する。【解決手段】電子データ履歴から情報を取得する手段と、取得した情報を解析して特徴を抽出する手段と、抽出した特徴に基づいて個別の生成モデルを生成する手段と、生成された生成モデルを用いて新たな情報に対する回答を自動生成する手段と、家庭内で使用される通信手段から得られるデータを解析する手段と、解析されたデータを基に家庭内メンバーに特化した助言を提供する手段と、前記回答の評価を受けて生成モデルを調整する手段と、を含むシステム。

Description

Technical Field

[0004] , , , ,

[0005] , , , ,

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In modern enterprises, the decline in business efficiency due to insufficient communication between departments has become an issue. Also, repetitive responses to the same questions and the loss of knowledge of personnel who have left the organization are important problems. These problems hinder the efficient performance of business by all employees and further lead to a decrease in the intellectual assets of the organization, so appropriate solutions are required.

Means for Solving the Problems

[0005] This invention provides a system that constructs digital clones of employees by generating individual data models from their electronic communication history and automatically generates responses through a chat tool. This enables efficient handling of similar questions. Furthermore, by applying natural language processing algorithms, it generates data models tailored to each employee's knowledge and style, providing highly accurate information. In addition, by retaining the communication history of employees who have left the organization, it is possible to prevent knowledge loss and promote knowledge sharing within the organization. Through the above means, it is possible to improve communication within companies and streamline knowledge management.

[0006] "Electronic communication history" refers to a record of messages sent and received via a computer network, including information such as the date and time, sender, recipient, and message content.

[0007] "Means of acquiring information" refers to methods of collecting electronic communication history and related data in a specific format and extracting them as a dataset that can be analyzed.

[0008] "Methods for analyzing information and extracting features" refer to techniques that process acquired data, identify important patterns and relationships, and identify metadata useful for specific purposes.

[0009] A "data model" is an algorithmic structure generated based on a specific dataset, used to produce predictions and responses to new data.

[0010] "Means for automatically generating responses" refers to technologies that use data models to automatically create appropriate answers to given questions or requests using computer programs.

[0011] "Response evaluation" refers to the process of analyzing the accuracy and relevance of the generated responses and recording them as feedback for improvement.

[0012] "Natural language processing algorithms" refer to a set of methods and techniques that enable computers to understand and process human language, and are used for analyzing and generating text data.

[0013] "Knowledge loss" refers to the loss of information and knowledge due to employees leaving an organization, and includes a reduction in important assets, such as content essential for performing business operations. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, a labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, a labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the 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.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0028] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention can be implemented as a system for streamlining internal company communication. Specific embodiments are described below.

[0036] The server periodically retrieves and stores the company's electronic communication history. Specifically, after the end of each workday, the server collects message data from all employees via the chat tool's API and stores it in a database. This data includes the message sender, recipient, content, and timestamp.

[0037] Next, the server analyzes the stored data. Using natural language processing algorithms, the server analyzes the message structure and extracts features that allow for interpretation of meaning and intent. This reveals what topics each employee is knowledgeable about and their response style.

[0038] Subsequently, the server generates digital clones of each employee from the analyzed data. These are individualized generative AI models that reflect each employee's communication style and expertise. The digital clones are then adjusted to enable nuanced responses using past conversational context and expertise.

[0039] On the other hand, users can send questions and inquiries that arise in their daily work via a chat tool from their device. For example, if they want to check the progress of a new project, they can send the question to a digital clone of the person in charge.

[0040] Upon receiving this question, the server automatically generates an appropriate answer using the corresponding digital clone. The server then sends the generated answer back to the terminal for display to the user. Because this answer includes knowledge that acts as a liaison between departments, accurate information is provided even for questions concerning the work of other departments.

[0041] Furthermore, users can evaluate the quality of the responses and input feedback into their devices. This feedback is collected by the server and used to further improve the accuracy of the digital clones. This allows the model to continuously improve, and knowledge management within the company is continuously optimized.

[0042] In this way, the present invention is implemented to improve the work efficiency of employees while simultaneously strengthening knowledge sharing throughout the organization.

[0043] The following describes the processing flow.

[0044] Step 1:

[0045] The server retrieves electronic communication history via the company's internal chat tool API at specific times and stores it in a database. This process collects data including the message sender, recipient, content, and timestamp.

[0046] Step 2:

[0047] The server preprocesses the stored chat data. It cleans the text, tokenizes it into a format that is easy for the language model to handle, and removes data noise.

[0048] Step 3:

[0049] The server uses pre-processed data to run natural language processing algorithms and extract key features from the messages. These include frequently occurring terms and each employee's response patterns.

[0050] Step 4:

[0051] The server generates individual digital clone models from the extracted features. These models reflect each employee's expertise and communication style.

[0052] Step 5:

[0053] Users send questions via a chat tool from their device. For example, they might type a question about the progress of a new project.

[0054] Step 6:

[0055] The server receives questions from users and automatically generates answers using the corresponding digital clones.

[0056] Step 7:

[0057] The server sends the generated response to the terminal and displays it to the user. The user can then proceed with their work based on this response.

[0058] Step 8:

[0059] Users verify the accuracy of the answers and enter feedback into their devices. For example, they might rate the answer as "partially correct."

[0060] Step 9:

[0061] The server receives feedback from users and uses it to improve the digital clone model. This allows the model to continue learning and improve its response accuracy.

[0062] (Example 1)

[0063] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0064] In modern organizations, improving the efficiency of internal communication and information sharing are crucial challenges. However, employee knowledge tends to remain tacit and confined to individuals, and important knowledge can be lost due to employee turnover or transfers. Furthermore, to answer questions that arise in daily work quickly and accurately, responses that reflect each employee's work knowledge and communication style are necessary. However, automating this process has been difficult with traditional methods.

[0065] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0066] In this invention, the server includes means for collecting and storing the history of electronic messages, means for analyzing the stored data using natural language processing and extracting features, and means for creating individual generative AI models based on the extracted features. This allows for the construction of individual AI models that reflect each employee's expertise and communication style, enabling the provision of quick and accurate automated responses to user questions. Furthermore, since the AI ​​models can be continuously improved through evaluation of the responses, internal knowledge management and communication efficiency are enhanced.

[0067] "Electronic message history" refers to the record of all messages exchanged through the organization's internal communication channels. This record includes details such as the sender, recipient, content, and timestamp.

[0068] "Natural language processing" refers to the technology used to process human language using computers, and is a means of extracting information from text by performing grammatical analysis and semantic analysis.

[0069] "Features" refer to important attributes and characteristics of information extracted from the content and structure of a message, and these are used to perform more accurate analysis and response generation.

[0070] A "generative AI model" refers to an artificial intelligence model that is trained to automatically analyze information and generate responses for specific tasks.

[0071] A "prompt" refers to an instruction or question that a user inputs to guide a generated AI model.

[0072] "Automated response" refers to a response that is mechanically generated using a generative AI model in response to a user's prompt.

[0073] "Evaluation" refers to the quality assessment conducted by users regarding the accuracy and usefulness of the generated automated responses, and the results of this evaluation are used to improve the AI ​​model.

[0074] This invention can be implemented as a system that streamlines communication within an organization and enables smooth information sharing. This system operates through the collaboration of a server, terminals, and users, with the server playing a central role in performing advanced data analysis and response generation.

[0075] The server first automatically collects communication history from the organization's electronic messaging platform via an API and stores this data in a secure database. This data includes the sender, recipient, content, and timestamp of each message. The server then applies natural language processing algorithms to the stored data to extract important features and patterns from the message text. This analysis is expected to utilize Python natural language processing libraries such as NLTK or spaCy.

[0076] After features are extracted, the server generates generative AI models corresponding to individual employees or members. These generative AI models reflect each employee's communication style and expertise based on their past messaging history. The AI ​​is trained using machine learning libraries such as TENSORFLOW® and PyTorch to generate these models.

[0077] Users can send questions and inquiries related to their daily work to the generated AI model as prompts via their own devices. An example of a prompt a user might send is, "Please tell me the details of the new marketing strategy."

[0078] Based on the prompt received from the user, the server generates an appropriate automated response using the corresponding generative AI model. The generated response is immediately sent back to the terminal and displayed to the user. This process allows the user to obtain information quickly and accurately.

[0079] Furthermore, users can input feedback by evaluating each response. This feedback is collected by the server and used for the continuous improvement of the generated AI model. Through these features, the system will constantly improve and optimize knowledge and information management across the organization. It is also expected that the accuracy of responses to specific prompts will improve, leading to increased employee productivity.

[0080] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0081] Step 1:

[0082] The server periodically collects the history of electronic messages from the organization's electronic messaging platform via an API. The input is message data retrieved from the platform. Specifically, the server runs an automated script daily after the end of the workday to extract message data. This includes the sender, recipient, content, and timestamp for each message. This data is stored in a database for further processing. The output is the structured message data stored in the database.

[0083] Step 2:

[0084] The server analyzes stored message data using natural language processing algorithms. The input is message data from a database. Specifically, the server uses Python's natural language processing library to perform text analysis, including morphological and grammatical analysis. As a result of this analysis, important keywords and phrases are extracted. The output is the extracted feature data.

[0085] Step 3:

[0086] The server creates individual generative AI models based on extracted feature data. The input is the feature data. The server uses machine learning libraries to train generative AI models that reflect each employee's communication style and expertise. This training includes a process of adjusting the model parameters by referring to past messages. The output is the generative AI model for each employee.

[0087] Step 4:

[0088] The user uses their device to send a question as a prompt to the generating AI model. The input is a specific prompt generated by the user. A concrete example would be a question like, "Please tell me the details of the new marketing strategy." The user's input is sent from the device to the server.

[0089] Step 5:

[0090] The server automatically generates a response using the appropriate generative AI model based on the received prompt. The input consists of the prompt and the generative AI model. The server analyzes the prompt and applies it to the AI ​​model to generate an appropriate answer. During this process, the generative AI model constructs an accurate answer by referencing past data. The output is the generated response message.

[0091] Step 6:

[0092] The terminal displays the response received from the server to the user. The input is the response message sent from the server. The terminal formats the received content appropriately and displays it on the user's screen. The output is the response in a format viewable by the user.

[0093] Step 7:

[0094] Users evaluate the quality of the generated responses and enter feedback into the device. This feedback consists of user ratings and comments. Using a feedback form on the device, users provide their opinions on the accuracy and satisfaction level of the responses.

[0095] Step 8:

[0096] The server collects user feedback and uses it to improve the generative AI model. The input is user feedback. The server analyzes the feedback data to help retrain the AI ​​model. This process improves the model's performance and enhances future response accuracy. The output is the improved generative AI model.

[0097] (Application Example 1)

[0098] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0099] Insufficient information sharing among family members can lead to misunderstandings and inefficient communication. Furthermore, it is difficult to accommodate differing communication styles within the family, and there is a lack of means to provide advice tailored to individual needs. It is necessary to address these challenges and achieve smooth communication and information sharing within the family.

[0100] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0101] In this invention, the server includes means for acquiring information from electronic data history, means for analyzing the acquired information and extracting features, means for generating individual generative models based on the extracted features, means for analyzing data obtained from communication means used within the home, and means for providing advice tailored to each member of the household based on the analyzed data. This facilitates information sharing among members of the household and enables effective communication tailored to each member.

[0102] "Electronic data history" refers to a collection of digital information that records past communications and information exchanges within a household or organization.

[0103] "Methods for extracting features" refer to techniques and methods for finding important patterns and trends from acquired data and extracting information that forms the basis of analysis.

[0104] A "generative model" is a digital model that can generate or respond to information based on analyzed features, depending on a specific task or situation.

[0105] "Communication methods" refer to a group of devices and applications used to exchange information in the form of voice, text, video, etc.

[0106] "Means of providing advice" refers to methods and systems that use collected data and generative models to provide appropriate suggestions and information to individual users.

[0107] The embodiment of this invention is based on a system that streamlines communication within the home and promotes information sharing. The server periodically collects and stores the history of electronic data used within the home. Specifically, it aggregates the history obtained from various communication methods used within the home (e.g., messaging apps on smartphones and voice commands via smart speakers) and stores it in a database.

[0108] Next, the server applies natural language processing algorithms to the stored data to analyze each member's communication style, frequently occurring topics, and intentions. Based on the analysis results, it constructs a generative AI model tailored to each member. This generative AI model utilizes advanced algorithms such as OpenAI's GPT-3, enabling nuanced communication that takes into account each member's past conversational context.

[0109] Users send questions and requests for information that arise in their daily lives to the server via their smart devices. For example, if a user wants to plan family activities for the next weekend, they might ask their smart speaker for their schedule. The server, upon receiving this question, automatically generates appropriate suggestions and answers based on the AI ​​model of the relevant family member and sends them back to the device. This advice and answers can also be used as part of information sharing for the entire family, and are designed to make it easy for all members to understand the related content.

[0110] For example, if a parent asks a smart speaker, "Tell me some good places for a family picnic next weekend," the system can suggest the best location based on past picnic history and the family's preferences. An example of a prompt used in this case would be, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0111] In this way, the system can improve the overall efficiency of information transfer within the household by facilitating the integration of knowledge and smooth communication, and by providing optimized advice tailored to each individual member.

[0112] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0113] Step 1:

[0114] The server collects the history of electronic data used within the home. Specifically, it periodically retrieves message and voice command history data from communication devices such as smartphones and smart speakers and stores it in a database. The input to this process is raw data from the devices, and the output is an organized history of electronic data.

[0115] Step 2:

[0116] The server applies natural language processing algorithms to the collected electronic data history. In this process, features are extracted from the data, and the communication styles and frequently occurring topics of each household member are analyzed. The input is the organized electronic data history, and the output is the data from which the features have been extracted. Specifically, algorithms such as OpenAI's GPT-3 are used to model the dialogue patterns of each member.

[0117] Step 3:

[0118] The server builds a generative AI model tailored to each household member based on the analyzed features. This generative AI model enables communication simulations that take past conversational context into account. The input is the data from which features have been extracted, and the output is the constructed generative AI model.

[0119] Step 4:

[0120] Users input questions and requests for information to the server via their smart devices. For example, they might tell a smart speaker, "Can you recommend a good picnic spot for next weekend?"

[0121] Step 5:

[0122] The server generates appropriate answers using a generative AI model based on the received request. This process uses historical data and prompts to process information and create relevant information tailored to the user. The input is the user's question, and the output is the generated answer. An example of a prompt is, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0123] Step 6:

[0124] The terminal displays or guides the user with the response received from the server. This allows the user to review the response proposed by the generating AI model and obtain the necessary information. The input for this step is the generated response, and the output is the information provided to the user.

[0125] Step 7:

[0126] Users evaluate the quality of the information and answers provided and send feedback back to the server. This feedback is used to improve the accuracy of the generative AI model. The input is user feedback, and the output is the data that is used to improve accuracy.

[0127] This series of processes effectively optimizes information sharing and communication within the household.

[0128] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0129] This invention is a system that analyzes electronic communication history to generate a digital clone and further combines it with an emotion engine to provide flexible responses based on the user's emotions. Specific embodiments of this invention are described below.

[0130] First, the server periodically retrieves the company's electronic communication history. This includes past message data, sender, recipient, and timestamp. For example, the server collects communication data daily and builds a dataset for each employee.

[0131] Next, the server analyzes the acquired data. Using natural language processing algorithms, it extracts text features and generates a digital clone of each employee based on these features. This digital clone is an individual data model that reflects the communication style and knowledge in the workplace.

[0132] The emotion engine analyzes acquired electronic communication data to identify emotions from messages. For example, the server detects signs of positive, negative, or neutral emotions from specific word choices and expressions.

[0133] When a user submits a question to the server, the server automatically generates a response using a corresponding digital clone. At this point, an emotion engine intervenes, adjusting the response based on the user's current emotional state. For example, if an emotion indicating dissatisfaction or stress is detected, the server will create a response in a more considerate tone.

[0134] The generated response is sent to the terminal and displayed to the user. The user can review the response and provide feedback as needed. This feedback is used by the server to improve the emotion engine and digital clone model.

[0135] For example, if a user asks a question with a nuance like "I'm worried about the recent project," the server will consider the emotion detected by the emotion engine and generate a considerate response such as, "That's a concern. I'll check on the project's progress and suggest ways to improve it."

[0136] This invention enables efficient and emotionally sensitive communication within a company. This, in turn, leads to smoother business operations and improved communication within the organization.

[0137] The following describes the processing flow.

[0138] Step 1:

[0139] The server periodically retrieves communication history via the company's internal electronic communication tools' APIs and stores it in a database. The stored data includes sender, recipient, message text, and timestamp.

[0140] Step 2:

[0141] The server preprocesses the stored data. It formats it into a specific format, performs tokenization and semantic analysis using natural language processing algorithms, and prepares the dataset.

[0142] Step 3:

[0143] The server generates digital clones of each employee based on the prepared dataset. Each clone is an individual model used to generate responses, reflecting the employee's unique communication style and expertise.

[0144] Step 4:

[0145] The emotion engine performs text analysis on electronic communication data to identify emotions such as positive, negative, and neutral. The emotion engine determines emotions based on specific keywords and sentence structure.

[0146] Step 5:

[0147] Users input questions or requests from their terminals and send them to the server. For example, they might ask specific questions such as, "How can I overcome my concerns about a new project?"

[0148] Step 6:

[0149] The server generates a response based on the received question, utilizing the corresponding digital clone. It then adjusts the response to reflect the user's emotions, taking into account the results of the emotion engine's analysis.

[0150] Step 7:

[0151] The server sends back the generated response to the terminal. This response includes an adaptive tone and content that takes emotions into consideration.

[0152] Step 8:

[0153] Users review the responses displayed on their devices and provide feedback on whether the content is appropriate. For example, they might rate the response as helpful.

[0154] Step 9:

[0155] The server adjusts and improves the digital clone model and emotion engine based on user feedback. This maintains improved response accuracy and emotional adaptability for subsequent interactions.

[0156] (Example 2)

[0157] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0158] In today's business environment, communication via electronic means is commonplace. However, conventional systems are insufficient in generating responses that take user emotions into account, hindering effective communication. Furthermore, while providing responses that address emotional needs in internal corporate communication is crucial, the technology to achieve this is lacking. There is a need for highly accurate communication systems that can adjust to user emotions.

[0159] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0160] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, and means for generating individual data models based on the extracted features. This makes it possible to accurately determine the user's emotional state and generate a response adjusted accordingly.

[0161] "Electronic communication history" refers to records of messages, emails, chats, etc., sent and received through a company's internal or external communication network, and includes sender, recipient, timestamp, and message content.

[0162] "Means of acquiring information" refers to the process by which a server collects necessary communication history data from databases and storage on the network.

[0163] "Methods for extracting features" refer to using natural language processing techniques to analyze acquired electronic communication history data and reveal text features such as language usage frequency and specific grammatical patterns.

[0164] An "individualized data model" refers to a digital personality model that reflects each user's communication style and business knowledge, and is a formalized representation of the user's characteristics.

[0165] "Means for automatically generating responses" refers to technologies in which a system automatically creates a response to user input by utilizing a generated data model.

[0166] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of digital clone models and their response generation based on user feedback.

[0167] "Means for determining emotional state" refers to algorithms that analyze text data extracted from communication history to identify emotions such as positive, negative, and neutral.

[0168] "Means of adjusting tone and content" refers to the process of appropriately modifying the wording and content of generated responses to match the emotional state of the identified user.

[0169] This invention is a system that generates a digital clone of a user using their electronic communication history within a company and provides responses that take their emotional state into consideration. An embodiment of this system is shown below.

[0170] First, the server is connected to the company's internal network and periodically retrieves electronic communication history from mail servers and messaging services. This data includes sender, recipient, timestamp, and message content. The server aggregates this data to build a communication history database for each employee.

[0171] Next, this system utilizes natural language processing (NLP) software. The server analyzes the collected communication history data and extracts linguistic features. This extraction process analyzes the grammar, structure, and frequency of terms in the text. Based on this, a digital clone is generated for each employee. Each digital clone is an individual data model that reflects each employee's communication style and work knowledge.

[0172] Furthermore, using an emotion engine, the server determines the user's emotional state from the communication history. It analyzes message text to identify emotions such as positive, negative, or neutral. For example, it infers the user's current emotions from specific words and expressions and adjusts its response accordingly.

[0173] When a user sends a question or request to the server using their device, the server generates a response using an appropriate digital clone. In this process, the server adjusts the response based on information from the emotion engine, ensuring that communication is sensitive to the user's feelings. For example, in response to a user expressing concern about project progress, the server might create a considerate response such as, "Don't worry. We'll check the progress and consider solutions."

[0174] The generated response is sent to the user's device, allowing the user to review the displayed information. Users can also provide feedback on the response, which the server uses to improve the emotion engine and digital clone.

[0175] As a concrete example, a prompt message could be, "Use a digital clone of an employee to generate a thoughtful response based on the user's emotions." This system is expected to improve the quality and efficiency of communication within the company.

[0176] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0177] Step 1:

[0178] The server periodically retrieves electronic communication history from the corporate network. Input includes sender, recipient, timestamp, and message content accessed from mail servers and messaging applications. The server aggregates this data to build a communication history database for each employee. This process results in the output of a digitized communication history for all employees.

[0179] Step 2:

[0180] The server uses natural language processing (NLP) software to analyze the acquired electronic communication history data. The input is the communication history database constructed in Step 1. Using this framework, the server extracts features such as text grammar, terminology, and frequency, and generates digital clones that reflect the user's communication style and business knowledge. The output is the digital clones as individual data models.

[0181] Step 3:

[0182] The server utilizes an emotion engine to determine the user's emotional state from communication history data. The inputs are the digital clone model generated in step 2 and the text data of the electronic communication history. Here, the wording and expressions in the text are analyzed to identify emotions such as positive, negative, and neutral, and the user's emotional state is quantified. The output of this process is the user's current emotional state data.

[0183] Step 4:

[0184] Users can ask questions and make requests to the server using a terminal. The input is a query sent by the user. After receiving this query, the server automatically generates the optimal response, taking into account the digital clone from step 2 and the emotional state from step 3. Specifically, the server adjusts the tone and content of the response based on the user's emotions. The output is the adjusted response message sent to the user.

[0185] Step 5:

[0186] The user reviews the response displayed on the terminal. The user can submit feedback on the response. The input is the user's feedback data. The server receives this feedback and integrates it into the digital clone model and emotion engine, performing an improvement process to enhance the accuracy of future response generation. The output is the improved system behavior.

[0187] (Application Example 2)

[0188] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0189] In today's digital society, electronic communication has become an everyday means of communication. However, inhuman responses and misunderstandings in electronic communication can sometimes degrade the quality of individual communication. Furthermore, within family settings and organizations, there is a need for means to appropriately recognize individual emotions and facilitate smooth relationships. Therefore, a system that utilizes electronic communication history to generate responses adapted to emotions is desired.

[0190] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0191] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, means for generating individual data models based on the extracted features, means for automatically generating responses to new information using the generated data models, means for adjusting the generated responses to take into account the user's emotional state, and means for modifying the data models based on evaluations of the responses. This enables flexible communication that reflects the user's emotions.

[0192] "Electronic communication history" is a general term for the records of emails and messages sent and received by individual users.

[0193] "Means of acquiring information" refers to the function that allows a server to collect necessary data from external databases or networks.

[0194] "Means of analyzing information" refers to the process of analyzing acquired data in detail and extracting important features.

[0195] "Methods for extracting features" refer to techniques that identify useful information from analyzed data and perform analysis based on that information.

[0196] An "individualized data model" is a digital representation that reflects the communication style and emotions of a specific user.

[0197] "Means for automatically generating responses" refers to a method of mechanically creating an appropriate response based on the input information.

[0198] "Means of adjusting based on emotional state" refers to technologies that identify the user's emotional response and modify the content and tone of the response accordingly.

[0199] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of the model and the quality of responses by incorporating feedback from users.

[0200] The system implementing this invention functions as a household robot. A server acquires electronic communication history to form a specific dataset. The acquired data is analyzed using a natural language processing algorithm, which generates individual digital clones. These clones reflect the user's past communication style and emotional tendencies.

[0201] Furthermore, the server uses a sentiment analysis engine to analyze the emotions contained in user inquiries. The sentiment analysis software used here is the "sentiment_analysis" library. Based on the analyzed sentiment data, the tone and content of the response are adjusted. In this process, the "openai" API is utilized to automatically generate appropriate responses.

[0202] The generated response is delivered to the user through a home robot. The user can review the response and provide additional feedback. This feedback is collected on a server and used to improve the accuracy of the digital clone model.

[0203] For example, if a user comments, "Work is tough and I'm tired," a home robot can offer a thoughtful suggestion such as, "How about we watch a relaxing movie together today?" In this case, an example of a prompt for the generative AI model would be:

[0204] "Emotion: Considerate tone. Question: 'Work is tough and I'm tired.' Response:"

[0205] In this way, home robots provide an environment that promotes better communication based on the user's emotions.

[0206] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0207] Step 1:

[0208] The server retrieves electronic communication history from a database. The input is historical data of past emails and messages. Based on this data, it formats information such as date, sender, and recipient to construct a dataset. The output is a processed dataset corresponding to each user.

[0209] Step 2:

[0210] The server uses natural language processing algorithms to analyze the acquired dataset. The input is the dataset formatted in step 1, and this data is subjected to text analysis to extract features. As a result, it outputs digital clones that model each user's communication patterns and emotional tendencies.

[0211] Step 3:

[0212] The terminal (a home robot) uses an emotion analysis engine to receive real-time questions and comments from the user. The input is the user's current inquiry, and emotion analysis detects the emotional tone of its content. The output is the detected emotion data.

[0213] Step 4:

[0214] The server automatically generates responses based on digital clones and emotion data. The input consists of the digital clones from step 2 and the emotion data from step 3, and it generates responses using appropriate prompts that reflect the emotions via the "openai" API. The output is a tailored response that takes the user's emotions into consideration.

[0215] Step 5:

[0216] The device presents the generated response to the user. The input is the response generated in step 4, and interactive communication takes place by conveying this to the user via voice or text. The output is the user's feedback in the form of words or actions.

[0217] Step 6:

[0218] The server collects user feedback and uses it to update the digital clone and response generation algorithms. The input is the user feedback obtained in step 5, and analyzing this feedback and adjusting the model improves the accuracy and appropriateness of the response. The output is the improved digital clone model.

[0219] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0220] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0221] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0222] [Second Embodiment]

[0223] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0224] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0225] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0226] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0227] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0228] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0229] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0230] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0231] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0232] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0233] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0234] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0235] This invention can be implemented as a system for streamlining internal company communication. Specific embodiments are described below.

[0236] The server periodically retrieves and stores the company's electronic communication history. Specifically, after the end of each workday, the server collects message data from all employees via the chat tool's API and stores it in a database. This data includes the message sender, recipient, content, and timestamp.

[0237] Next, the server analyzes the stored data. Using natural language processing algorithms, the server analyzes the message structure and extracts features that allow for interpretation of meaning and intent. This reveals what topics each employee is knowledgeable about and their response style.

[0238] Subsequently, the server generates digital clones of each employee from the analyzed data. These are individualized generative AI models that reflect each employee's communication style and expertise. The digital clones are then adjusted to enable nuanced responses using past conversational context and expertise.

[0239] On the other hand, users can send questions and inquiries that arise in their daily work via a chat tool from their device. For example, if they want to check the progress of a new project, they can send the question to a digital clone of the person in charge.

[0240] Upon receiving this question, the server automatically generates an appropriate answer using the corresponding digital clone. The server then sends the generated answer back to the terminal for display to the user. Because this answer includes knowledge that acts as a liaison between departments, accurate information is provided even for questions concerning the work of other departments.

[0241] Furthermore, users can evaluate the quality of the responses and input feedback into their devices. This feedback is collected by the server and used to further improve the accuracy of the digital clones. This allows the model to continuously improve, and knowledge management within the company is continuously optimized.

[0242] In this way, the present invention is implemented to improve the work efficiency of employees while simultaneously strengthening knowledge sharing throughout the organization.

[0243] The following describes the processing flow.

[0244] Step 1:

[0245] The server retrieves electronic communication history via the company's internal chat tool API at specific times and stores it in a database. This process collects data including the message sender, recipient, content, and timestamp.

[0246] Step 2:

[0247] The server preprocesses the stored chat data. It cleans the text, tokenizes it into a format that is easy for the language model to handle, and removes data noise.

[0248] Step 3:

[0249] The server uses pre-processed data to run natural language processing algorithms and extract key features from the messages. These include frequently occurring terms and each employee's response patterns.

[0250] Step 4:

[0251] The server generates individual digital clone models from the extracted features. These models reflect each employee's expertise and communication style.

[0252] Step 5:

[0253] Users send questions via a chat tool from their device. For example, they might type a question about the progress of a new project.

[0254] Step 6:

[0255] The server receives questions from users and automatically generates answers using the corresponding digital clones.

[0256] Step 7:

[0257] The server sends the generated response to the terminal and displays it to the user. The user can then proceed with their work based on this response.

[0258] Step 8:

[0259] Users verify the accuracy of the answers and enter feedback into their devices. For example, they might rate the answer as "partially correct."

[0260] Step 9:

[0261] The server receives feedback from users and uses it to improve the digital clone model. This allows the model to continue learning and improve its response accuracy.

[0262] (Example 1)

[0263] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0264] In modern organizations, improving the efficiency of internal communication and information sharing are crucial challenges. However, employee knowledge tends to remain tacit and confined to individuals, and important knowledge can be lost due to employee turnover or transfers. Furthermore, to answer questions that arise in daily work quickly and accurately, responses that reflect each employee's work knowledge and communication style are necessary. However, automating this process has been difficult with traditional methods.

[0265] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0266] In this invention, the server includes means for collecting and storing the history of electronic messages, means for analyzing the stored data using natural language processing and extracting features, and means for creating individual generative AI models based on the extracted features. This allows for the construction of individual AI models that reflect each employee's expertise and communication style, enabling the provision of quick and accurate automated responses to user questions. Furthermore, since the AI ​​models can be continuously improved through evaluation of the responses, internal knowledge management and communication efficiency are enhanced.

[0267] "Electronic message history" refers to the record of all messages exchanged through the organization's internal communication channels. This record includes details such as the sender, recipient, content, and timestamp.

[0268] "Natural language processing" refers to the technology used to process human language using computers, and is a means of extracting information from text by performing grammatical analysis and semantic analysis.

[0269] "Features" refer to important attributes and characteristics of information extracted from the content and structure of a message, and these are used to perform more accurate analysis and response generation.

[0270] A "generative AI model" refers to an artificial intelligence model that is trained to automatically analyze information and generate responses for specific tasks.

[0271] A "prompt" refers to an instruction or question that a user inputs to guide a generated AI model.

[0272] "Automated response" refers to a response that is mechanically generated using a generative AI model in response to a user's prompt.

[0273] "Evaluation" refers to the quality assessment conducted by users regarding the accuracy and usefulness of the generated automated responses, and the results of this evaluation are used to improve the AI ​​model.

[0274] This invention can be implemented as a system that streamlines communication within an organization and enables smooth information sharing. This system operates through the collaboration of a server, terminals, and users, with the server playing a central role in performing advanced data analysis and response generation.

[0275] The server first automatically collects communication history from the organization's electronic messaging platform via an API and stores this data in a secure database. This data includes the sender, recipient, content, and timestamp of each message. The server then applies natural language processing algorithms to the stored data to extract important features and patterns from the message text. This analysis is expected to utilize Python natural language processing libraries such as NLTK or spaCy.

[0276] After features are extracted, the server generates generative AI models corresponding to individual employees or members. These generative AI models reflect each employee's communication style and expertise based on their past messaging history. The AI ​​is trained using machine learning libraries such as TensorFlow and PyTorch to generate these models.

[0277] Users can send questions and inquiries related to their daily work to the generated AI model as prompts via their own devices. An example of a prompt a user might send is, "Please tell me the details of the new marketing strategy."

[0278] Based on the prompt received from the user, the server generates an appropriate automated response using the corresponding generative AI model. The generated response is immediately sent back to the terminal and displayed to the user. This process allows the user to obtain information quickly and accurately.

[0279] Furthermore, the user can input an evaluation of each response as feedback. This feedback is collected by the server and used for the continuous improvement of the generative AI model. Through such functions, this system is constantly improved, optimizing knowledge and information management across the organization. Also, it is expected that the accuracy of responses to specific prompt texts will be improved, enhancing the labor productivity of employees.

[0280] The flow of the specific process in Example 1 will be described using FIG. 11.

[0281] Step 1:

[0282] The server periodically collects the history of electronic messages through an API from the electronic messaging platform within the organization. The input is the message data obtained from the platform. As a specific operation, the server executes an automatic script after work every day to extract the message data. This includes the sender, recipient, content, and timestamp of each message. This data is stored in a database for further processing. The output is the structured message data stored in the database.

[0283] Step 2:

[0284] The server analyzes the stored message data using natural language processing algorithms. The input is the message data in the database. As a specific operation, the server uses a natural language processing library in Python to perform text analysis and conduct morphological and syntactic analysis. As a result of this analysis, important keywords and phrases are extracted. The output is the extracted feature data.

[0285] Step 3:

[0286] The server creates an individual generative AI model based on the extracted feature data. The input is the feature data. The server utilizes a machine learning library to train a generative AI model that reflects the communication style and expertise of each employee. This training includes a process of adjusting the model's parameters while referring to past messages. The output is a generative AI model for each employee.

[0287] Step 4:

[0288] The user uses their terminal to send a question as a prompt sentence to the generative AI model. The input is the specific prompt sentence generated by the user. As a specific example, inquiries such as "Please tell me the details of the new marketing strategy" can be considered. The user's input is sent from the terminal to the server.

[0289] Step 5:

[0290] Based on the received prompt sentence, the server automatically generates a response using the corresponding generative AI model. The input is the prompt sentence and the generative AI model. The server analyzes the prompt sentence and applies it to the AI model to generate an appropriate answer. In this process, the generative AI model constructs an accurate answer while referring to past data. The output is the generated response message.

[0291] Step 6:

[0292] The terminal displays the response received from the server to the user. The input is the response message sent from the server. The terminal formats the received content appropriately and displays it on the user's screen. The output is a response in a format that the user can view.

[0293] Step 7:

[0294] Users evaluate the quality of the generated responses and enter feedback into the device. This feedback consists of user ratings and comments. Using a feedback form on the device, users provide their opinions on the accuracy and satisfaction level of the responses.

[0295] Step 8:

[0296] The server collects user feedback and uses it to improve the generative AI model. The input is user feedback. The server analyzes the feedback data to help retrain the AI ​​model. This process improves the model's performance and enhances future response accuracy. The output is the improved generative AI model.

[0297] (Application Example 1)

[0298] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0299] Insufficient information sharing among family members can lead to misunderstandings and inefficient communication. Furthermore, it is difficult to accommodate differing communication styles within the family, and there is a lack of means to provide advice tailored to individual needs. It is necessary to address these challenges and achieve smooth communication and information sharing within the family.

[0300] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0301] In this invention, the server includes means for obtaining information from the electronic data history, means for analyzing the obtained information to extract features, means for generating an individual generation model based on the extracted features, means for analyzing data obtained from communication means used within the home, and means for providing advice specialized for household members based on the analyzed data. Thereby, it becomes possible to promote information sharing among household members and enable effective communication tailored to each member.

[0302] The "electronic data history" is a collection of digital information in which past communications and information exchanges within the home or organization are recorded.

[0303] The "means for extracting features" is a technique or method for finding important patterns and trends from the obtained data and extracting information that serves as the basis for analysis.

[0304] The "generation model" is a digital model that can generate information or provide responses according to specific tasks or situations based on the analyzed features.

[0305] The "communication means" is a group of devices and applications used for exchanging information in forms such as voice, text, video, etc.

[0306] The "means for providing advice" is a method or system for providing appropriate suggestions and information to individual users using the collected data and the generation model.

[0307] The embodiment for implementing this invention is based on a system that improves in-home communication and promotes information sharing. The server periodically collects the electronic data history used within the home and stores the data. Specifically, it aggregates the histories obtained from various communication means used within the home (e.g., message apps on smartphones and voice instructions by smart speakers) and stores them in a database.

[0308] Next, the server applies natural language processing algorithms to the stored data to analyze each member's communication style, frequently occurring topics, and intentions. Based on the analysis results, it builds a generative AI model tailored to each member. This generative AI model utilizes advanced algorithms such as OpenAI's GPT-3, enabling nuanced communication that takes into account each member's past conversational context.

[0309] Users send questions and requests for information that arise in their daily lives to the server via their smart devices. For example, if a user wants to plan family activities for the next weekend, they might ask their smart speaker for their schedule. The server, upon receiving this question, automatically generates appropriate suggestions and answers based on the AI ​​model of the relevant family member and sends them back to the device. This advice and answers can also be used as part of information sharing for the entire family, and are designed to make it easy for all members to understand the related content.

[0310] For example, if a parent asks a smart speaker, "Tell me some good places for a family picnic next weekend," the system can suggest the best location based on past picnic history and the family's preferences. An example of a prompt used in this case would be, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0311] In this way, the system can improve the overall efficiency of information transfer within the household by facilitating the integration of knowledge and smooth communication, and by providing optimized advice tailored to each individual member.

[0312] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0313] Step 1:

[0314] The server collects the history of electronic data used within the home. Specifically, it periodically retrieves message and voice command history data from communication devices such as smartphones and smart speakers and stores it in a database. The input to this process is raw data from the devices, and the output is an organized history of electronic data.

[0315] Step 2:

[0316] The server applies natural language processing algorithms to the collected electronic data history. In this process, features are extracted from the data, and the communication styles and frequently occurring topics of each household member are analyzed. The input is the organized electronic data history, and the output is the data from which the features have been extracted. Specifically, algorithms such as OpenAI's GPT-3 are used to model the dialogue patterns of each member.

[0317] Step 3:

[0318] The server builds a generative AI model tailored to each household member based on the analyzed features. This generative AI model enables communication simulations that take past conversational context into account. The input is the data from which features have been extracted, and the output is the constructed generative AI model.

[0319] Step 4:

[0320] Users input questions and requests for information to the server via their smart devices. For example, they might tell a smart speaker, "Can you recommend a good picnic spot for next weekend?"

[0321] Step 5:

[0322] The server generates appropriate answers using a generative AI model based on the received request. This process uses historical data and prompts to process information and create relevant information tailored to the user. The input is the user's question, and the output is the generated answer. An example of a prompt is, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0323] Step 6:

[0324] The terminal displays or guides the user with the response received from the server. This allows the user to review the response proposed by the generating AI model and obtain the necessary information. The input for this step is the generated response, and the output is the information provided to the user.

[0325] Step 7:

[0326] Users evaluate the quality of the information and answers provided and send feedback back to the server. This feedback is used to improve the accuracy of the generative AI model. The input is user feedback, and the output is the data that is used to improve accuracy.

[0327] This series of processes effectively optimizes information sharing and communication within the household.

[0328] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0329] This invention is a system that analyzes electronic communication history to generate a digital clone and further combines it with an emotion engine to provide flexible responses based on the user's emotions. Specific embodiments of this invention are described below.

[0330] First, the server periodically retrieves the company's electronic communication history. This includes past message data, sender, recipient, and timestamp. For example, the server collects communication data daily and builds a dataset for each employee.

[0331] Next, the server analyzes the acquired data. Using natural language processing algorithms, it extracts text features and generates a digital clone of each employee based on these features. This digital clone is an individual data model that reflects the communication style and knowledge in the workplace.

[0332] The emotion engine analyzes acquired electronic communication data to identify emotions from messages. For example, the server detects signs of positive, negative, or neutral emotions from specific word choices and expressions.

[0333] When a user submits a question to the server, the server automatically generates a response using a corresponding digital clone. At this point, an emotion engine intervenes, adjusting the response based on the user's current emotional state. For example, if an emotion indicating dissatisfaction or stress is detected, the server will create a response in a more considerate tone.

[0334] The generated response is sent to the terminal and displayed to the user. The user can review the response and provide feedback as needed. This feedback is used by the server to improve the emotion engine and digital clone model.

[0335] For example, if a user asks a question with a nuance like "I'm worried about the recent project," the server will consider the emotion detected by the emotion engine and generate a considerate response such as, "That's a concern. I'll check on the project's progress and suggest ways to improve it."

[0336] This invention enables efficient and emotionally sensitive communication within a company. This, in turn, leads to smoother business operations and improved communication within the organization.

[0337] The following describes the processing flow.

[0338] Step 1:

[0339] The server periodically retrieves communication history via the company's internal electronic communication tools' APIs and stores it in a database. The stored data includes sender, recipient, message text, and timestamp.

[0340] Step 2:

[0341] The server preprocesses the stored data. It formats it into a specific format, performs tokenization and semantic analysis using natural language processing algorithms, and prepares the dataset.

[0342] Step 3:

[0343] The server generates digital clones of each employee based on the prepared dataset. Each clone is an individual model used to generate responses, reflecting the employee's unique communication style and expertise.

[0344] Step 4:

[0345] The emotion engine performs text analysis on electronic communication data to identify emotions such as positive, negative, and neutral. The emotion engine determines emotions based on specific keywords and sentence structure.

[0346] Step 5:

[0347] Users input questions or requests from their terminals and send them to the server. For example, they might ask specific questions such as, "How can I overcome my concerns about a new project?"

[0348] Step 6:

[0349] The server generates a response based on the received question, utilizing the corresponding digital clone. It then adjusts the response to reflect the user's emotions, taking into account the results of the emotion engine's analysis.

[0350] Step 7:

[0351] The server sends back the generated response to the terminal. This response includes an adaptive tone and content that takes emotions into consideration.

[0352] Step 8:

[0353] Users review the responses displayed on their devices and provide feedback on whether the content is appropriate. For example, they might rate the response as helpful.

[0354] Step 9:

[0355] The server adjusts and improves the digital clone model and emotion engine based on user feedback. This maintains improved response accuracy and emotional adaptability for subsequent interactions.

[0356] (Example 2)

[0357] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0358] In today's business environment, communication via electronic means is commonplace. However, conventional systems are insufficient in generating responses that take user emotions into account, hindering effective communication. Furthermore, while providing responses that address emotional needs in internal corporate communication is crucial, the technology to achieve this is lacking. There is a need for highly accurate communication systems that can adjust to user emotions.

[0359] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0360] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, and means for generating individual data models based on the extracted features. This makes it possible to accurately determine the user's emotional state and generate a response adjusted accordingly.

[0361] "Electronic communication history" refers to records of messages, emails, chats, etc., sent and received through a company's internal or external communication network, and includes sender, recipient, timestamp, and message content.

[0362] "Means of acquiring information" refers to the process by which a server collects necessary communication history data from databases and storage on the network.

[0363] "Methods for extracting features" refer to using natural language processing techniques to analyze acquired electronic communication history data and reveal text features such as language usage frequency and specific grammatical patterns.

[0364] An "individualized data model" refers to a digital personality model that reflects each user's communication style and business knowledge, and is a formalized representation of the user's characteristics.

[0365] "Means for automatically generating responses" refers to technologies in which a system automatically creates a response to user input by utilizing a generated data model.

[0366] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of digital clone models and their response generation based on user feedback.

[0367] "Means for determining emotional state" refers to algorithms that analyze text data extracted from communication history to identify emotions such as positive, negative, and neutral.

[0368] "Means of adjusting tone and content" refers to the process of appropriately modifying the wording and content of generated responses to match the emotional state of the identified user.

[0369] This invention is a system that generates a digital clone of a user using their electronic communication history within a company and provides responses that take their emotional state into consideration. An embodiment of this system is shown below.

[0370] First, the server is connected to the company's internal network and periodically retrieves electronic communication history from mail servers and messaging services. This data includes sender, recipient, timestamp, and message content. The server aggregates this data to build a communication history database for each employee.

[0371] Next, this system utilizes natural language processing (NLP) software. The server analyzes the collected communication history data and extracts linguistic features. This extraction process analyzes the grammar, structure, and frequency of terms in the text. Based on this, a digital clone is generated for each employee. Each digital clone is an individual data model that reflects each employee's communication style and work knowledge.

[0372] Furthermore, using an emotion engine, the server determines the user's emotional state from the communication history. It analyzes message text to identify emotions such as positive, negative, or neutral. For example, it infers the user's current emotions from specific words and expressions and adjusts its response accordingly.

[0373] When a user sends a question or request to the server using their device, the server generates a response using an appropriate digital clone. In this process, the server adjusts the response based on information from the emotion engine, ensuring that communication is sensitive to the user's feelings. For example, in response to a user expressing concern about project progress, the server might create a considerate response such as, "Don't worry. We'll check the progress and consider solutions."

[0374] The generated response is sent to the user's device, allowing the user to review the displayed information. Users can also provide feedback on the response, which the server uses to improve the emotion engine and digital clone.

[0375] As a concrete example, a prompt message could be, "Use a digital clone of an employee to generate a thoughtful response based on the user's emotions." This system is expected to improve the quality and efficiency of communication within the company.

[0376] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0377] Step 1:

[0378] The server periodically retrieves electronic communication history from the corporate network. Input includes sender, recipient, timestamp, and message content accessed from mail servers and messaging applications. The server aggregates this data to build a communication history database for each employee. This process results in the output of a digitized communication history for all employees.

[0379] Step 2:

[0380] The server uses natural language processing (NLP) software to analyze the acquired electronic communication history data. The input is the communication history database constructed in Step 1. Using this framework, the server extracts features such as text grammar, terminology, and frequency, and generates digital clones that reflect the user's communication style and business knowledge. The output is the digital clones as individual data models.

[0381] Step 3:

[0382] The server utilizes an emotion engine to determine the user's emotional state from communication history data. The inputs are the digital clone model generated in step 2 and the text data of the electronic communication history. Here, the wording and expressions in the text are analyzed to identify emotions such as positive, negative, and neutral, and the user's emotional state is quantified. The output of this process is the user's current emotional state data.

[0383] Step 4:

[0384] Users can ask questions and make requests to the server using a terminal. The input is a query sent by the user. After receiving this query, the server automatically generates the optimal response, taking into account the digital clone from step 2 and the emotional state from step 3. Specifically, the server adjusts the tone and content of the response based on the user's emotions. The output is the adjusted response message sent to the user.

[0385] Step 5:

[0386] The user reviews the response displayed on the terminal. The user can submit feedback on the response. The input is the user's feedback data. The server receives this feedback and integrates it into the digital clone model and emotion engine, performing an improvement process to enhance the accuracy of future response generation. The output is the improved system behavior.

[0387] (Application Example 2)

[0388] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0389] In today's digital society, electronic communication has become an everyday means of communication. However, inhuman responses and misunderstandings in electronic communication can sometimes degrade the quality of individual communication. Furthermore, within family settings and organizations, there is a need for means to appropriately recognize individual emotions and facilitate smooth relationships. Therefore, a system that utilizes electronic communication history to generate responses adapted to emotions is desired.

[0390] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0391] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, means for generating individual data models based on the extracted features, means for automatically generating responses to new information using the generated data models, means for adjusting the generated responses to take into account the user's emotional state, and means for modifying the data models based on evaluations of the responses. This enables flexible communication that reflects the user's emotions.

[0392] "Electronic communication history" is a general term for the records of emails and messages sent and received by individual users.

[0393] "Means of acquiring information" refers to the function that allows a server to collect necessary data from external databases or networks.

[0394] "Means of analyzing information" refers to the process of analyzing acquired data in detail and extracting important features.

[0395] "Methods for extracting features" refer to techniques that identify useful information from analyzed data and perform analysis based on that information.

[0396] An "individualized data model" is a digital representation that reflects the communication style and emotions of a specific user.

[0397] "Means for automatically generating responses" refers to a method of mechanically creating an appropriate response based on the input information.

[0398] "Means of adjusting based on emotional state" refers to technologies that identify the user's emotional response and modify the content and tone of the response accordingly.

[0399] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of the model and the quality of responses by incorporating feedback from users.

[0400] The system implementing this invention functions as a household robot. A server acquires electronic communication history to form a specific dataset. The acquired data is analyzed using a natural language processing algorithm, which generates individual digital clones. These clones reflect the user's past communication style and emotional tendencies.

[0401] Furthermore, the server uses a sentiment analysis engine to analyze the emotions contained in user inquiries. The sentiment analysis software used here is the "sentiment_analysis" library. Based on the analyzed sentiment data, the tone and content of the response are adjusted. In this process, the "openai" API is utilized to automatically generate appropriate responses.

[0402] The generated response is delivered to the user through a home robot. The user can review the response and provide additional feedback. This feedback is collected on a server and used to improve the accuracy of the digital clone model.

[0403] For example, if a user comments, "Work is tough and I'm tired," a home robot can offer a thoughtful suggestion such as, "How about we watch a relaxing movie together today?" In this case, an example of a prompt for the generative AI model would be:

[0404] "Emotion: Considerate tone. Question: 'Work is tough and I'm tired.' Response:"

[0405] In this way, home robots provide an environment that promotes better communication based on the user's emotions.

[0406] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0407] Step 1:

[0408] The server retrieves electronic communication history from a database. The input is historical data of past emails and messages. Based on this data, it formats information such as date, sender, and recipient to construct a dataset. The output is a processed dataset corresponding to each user.

[0409] Step 2:

[0410] The server uses natural language processing algorithms to analyze the acquired dataset. The input is the dataset formatted in step 1, and this data is subjected to text analysis to extract features. As a result, it outputs digital clones that model each user's communication patterns and emotional tendencies.

[0411] Step 3:

[0412] The terminal (a home robot) uses an emotion analysis engine to receive real-time questions and comments from the user. The input is the user's current inquiry, and emotion analysis detects the emotional tone of its content. The output is the detected emotion data.

[0413] Step 4:

[0414] The server automatically generates responses based on digital clones and emotion data. The input consists of the digital clones from step 2 and the emotion data from step 3, and it generates responses using appropriate prompts that reflect the emotions via the "openai" API. The output is a tailored response that takes the user's emotions into consideration.

[0415] Step 5:

[0416] The device presents the generated response to the user. The input is the response generated in step 4, and interactive communication takes place by conveying this to the user via voice or text. The output is the user's feedback in the form of words or actions.

[0417] Step 6:

[0418] The server collects user feedback and uses it to update the digital clone and response generation algorithms. The input is the user feedback obtained in step 5, and analyzing this feedback and adjusting the model improves the accuracy and appropriateness of the response. The output is the improved digital clone model.

[0419] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0420] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0421] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0422] [Third Embodiment]

[0423] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0424] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0425] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0426] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0427] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0428] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0429] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0430] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0431] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0432] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0433] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0434] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0435] This invention can be implemented as a system for streamlining internal company communication. Specific embodiments are described below.

[0436] The server periodically retrieves and stores the company's electronic communication history. Specifically, after the end of each workday, the server collects message data from all employees via the chat tool's API and stores it in a database. This data includes the message sender, recipient, content, and timestamp.

[0437] Next, the server analyzes the stored data. Using natural language processing algorithms, the server analyzes the message structure and extracts features that allow for interpretation of meaning and intent. This reveals what topics each employee is knowledgeable about and their response style.

[0438] Subsequently, the server generates digital clones of each employee from the analyzed data. These are individualized generative AI models that reflect each employee's communication style and expertise. The digital clones are then adjusted to enable nuanced responses using past conversational context and expertise.

[0439] On the other hand, users can send questions and inquiries that arise in their daily work via a chat tool from their device. For example, if they want to check the progress of a new project, they can send the question to a digital clone of the person in charge.

[0440] Upon receiving this question, the server automatically generates an appropriate answer using the corresponding digital clone. The server then sends the generated answer back to the terminal for display to the user. Because this answer includes knowledge that acts as a liaison between departments, accurate information is provided even for questions concerning the work of other departments.

[0441] Furthermore, users can evaluate the quality of the responses and input feedback into their devices. This feedback is collected by the server and used to further improve the accuracy of the digital clones. This allows the model to continuously improve, and knowledge management within the company is continuously optimized.

[0442] In this way, the present invention is implemented to improve the work efficiency of employees while simultaneously strengthening knowledge sharing throughout the organization.

[0443] The following describes the processing flow.

[0444] Step 1:

[0445] The server retrieves electronic communication history via the company's internal chat tool API at specific times and stores it in a database. This process collects data including the message sender, recipient, content, and timestamp.

[0446] Step 2:

[0447] The server preprocesses the stored chat data. It cleans the text, tokenizes it into a format that is easy for the language model to handle, and removes data noise.

[0448] Step 3:

[0449] The server uses pre-processed data to run natural language processing algorithms and extract key features from the messages. These include frequently occurring terms and each employee's response patterns.

[0450] Step 4:

[0451] The server generates individual digital clone models from the extracted features. These models reflect each employee's expertise and communication style.

[0452] Step 5:

[0453] Users send questions via a chat tool from their device. For example, they might type a question about the progress of a new project.

[0454] Step 6:

[0455] The server receives questions from users and automatically generates answers using the corresponding digital clones.

[0456] Step 7:

[0457] The server sends the generated response to the terminal and displays it to the user. The user can then proceed with their work based on this response.

[0458] Step 8:

[0459] Users verify the accuracy of the answers and enter feedback into their devices. For example, they might rate the answer as "partially correct."

[0460] Step 9:

[0461] The server receives feedback from users and uses it to improve the digital clone model. This allows the model to continue learning and improve its response accuracy.

[0462] (Example 1)

[0463] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0464] In modern organizations, improving the efficiency of internal communication and information sharing are crucial challenges. However, employee knowledge tends to remain tacit and confined to individuals, and important knowledge can be lost due to employee turnover or transfers. Furthermore, to answer questions that arise in daily work quickly and accurately, responses that reflect each employee's work knowledge and communication style are necessary. However, automating this process has been difficult with traditional methods.

[0465] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0466] In this invention, the server includes means for collecting and storing the history of electronic messages, means for analyzing the stored data using natural language processing and extracting features, and means for creating individual generative AI models based on the extracted features. This allows for the construction of individual AI models that reflect each employee's expertise and communication style, enabling the provision of quick and accurate automated responses to user questions. Furthermore, since the AI ​​models can be continuously improved through evaluation of the responses, internal knowledge management and communication efficiency are enhanced.

[0467] "Electronic message history" refers to the record of all messages exchanged through the organization's internal communication channels. This record includes details such as the sender, recipient, content, and timestamp.

[0468] "Natural language processing" refers to the technology used to process human language using computers, and is a means of extracting information from text by performing grammatical analysis and semantic analysis.

[0469] "Features" refer to important attributes and characteristics of information extracted from the content and structure of a message, and these are used to perform more accurate analysis and response generation.

[0470] A "generative AI model" refers to an artificial intelligence model that is trained to automatically analyze information and generate responses for specific tasks.

[0471] A "prompt" refers to an instruction or question that a user inputs to guide a generated AI model.

[0472] "Automated response" refers to a response that is mechanically generated using a generative AI model in response to a user's prompt.

[0473] "Evaluation" refers to the quality assessment conducted by users regarding the accuracy and usefulness of the generated automated responses, and the results of this evaluation are used to improve the AI ​​model.

[0474] This invention can be implemented as a system that streamlines communication within an organization and enables smooth information sharing. This system operates through the collaboration of a server, terminals, and users, with the server playing a central role in performing advanced data analysis and response generation.

[0475] The server first automatically collects communication history from the organization's electronic messaging platform via an API and stores this data in a secure database. This data includes the sender, recipient, content, and timestamp of each message. The server then applies natural language processing algorithms to the stored data to extract important features and patterns from the message text. This analysis is expected to utilize Python natural language processing libraries such as NLTK or spaCy.

[0476] After features are extracted, the server generates generative AI models corresponding to individual employees or members. These generative AI models reflect each employee's communication style and expertise based on their past messaging history. The AI ​​is trained using machine learning libraries such as TensorFlow and PyTorch to generate these models.

[0477] Users can send questions and inquiries related to their daily work to the generated AI model as prompts via their own devices. An example of a prompt a user might send is, "Please tell me the details of the new marketing strategy."

[0478] Based on the prompt received from the user, the server generates an appropriate automated response using the corresponding generative AI model. The generated response is immediately sent back to the terminal and displayed to the user. This process allows the user to obtain information quickly and accurately.

[0479] Furthermore, users can input feedback by evaluating each response. This feedback is collected by the server and used for the continuous improvement of the generated AI model. Through these features, the system will constantly improve and optimize knowledge and information management across the organization. It is also expected that the accuracy of responses to specific prompts will improve, leading to increased employee productivity.

[0480] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0481] Step 1:

[0482] The server periodically collects the history of electronic messages from the organization's electronic messaging platform via an API. The input is message data retrieved from the platform. Specifically, the server runs an automated script daily after the end of the workday to extract message data. This includes the sender, recipient, content, and timestamp for each message. This data is stored in a database for further processing. The output is the structured message data stored in the database.

[0483] Step 2:

[0484] The server analyzes stored message data using natural language processing algorithms. The input is message data from a database. Specifically, the server uses Python's natural language processing library to perform text analysis, including morphological and grammatical analysis. As a result of this analysis, important keywords and phrases are extracted. The output is the extracted feature data.

[0485] Step 3:

[0486] The server creates individual generative AI models based on extracted feature data. The input is the feature data. The server uses machine learning libraries to train generative AI models that reflect each employee's communication style and expertise. This training includes a process of adjusting the model parameters by referring to past messages. The output is the generative AI model for each employee.

[0487] Step 4:

[0488] The user uses their device to send a question as a prompt to the generating AI model. The input is a specific prompt generated by the user. A concrete example would be a question like, "Please tell me the details of the new marketing strategy." The user's input is sent from the device to the server.

[0489] Step 5:

[0490] The server automatically generates a response using the appropriate generative AI model based on the received prompt. The input consists of the prompt and the generative AI model. The server analyzes the prompt and applies it to the AI ​​model to generate an appropriate answer. During this process, the generative AI model constructs an accurate answer by referencing past data. The output is the generated response message.

[0491] Step 6:

[0492] The terminal displays the response received from the server to the user. The input is the response message sent from the server. The terminal formats the received content appropriately and displays it on the user's screen. The output is the response in a format viewable by the user.

[0493] Step 7:

[0494] Users evaluate the quality of the generated responses and enter feedback into the device. This feedback consists of user ratings and comments. Using a feedback form on the device, users provide their opinions on the accuracy and satisfaction level of the responses.

[0495] Step 8:

[0496] The server collects user feedback and uses it to improve the generative AI model. The input is user feedback. The server analyzes the feedback data to help retrain the AI ​​model. This process improves the model's performance and enhances future response accuracy. The output is the improved generative AI model.

[0497] (Application Example 1)

[0498] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0499] Insufficient information sharing among family members can lead to misunderstandings and inefficient communication. Furthermore, it is difficult to accommodate differing communication styles within the family, and there is a lack of means to provide advice tailored to individual needs. It is necessary to address these challenges and achieve smooth communication and information sharing within the family.

[0500] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0501] In this invention, the server includes means for acquiring information from electronic data history, means for analyzing the acquired information and extracting features, means for generating individual generative models based on the extracted features, means for analyzing data obtained from communication means used within the home, and means for providing advice tailored to each member of the household based on the analyzed data. This facilitates information sharing among members of the household and enables effective communication tailored to each member.

[0502] "Electronic data history" refers to a collection of digital information that records past communications and information exchanges within a household or organization.

[0503] "Methods for extracting features" refer to techniques and methods for finding important patterns and trends from acquired data and extracting information that forms the basis of analysis.

[0504] A "generative model" is a digital model that can generate or respond to information based on analyzed features, depending on a specific task or situation.

[0505] "Communication methods" refer to a group of devices and applications used to exchange information in the form of voice, text, video, etc.

[0506] "Means of providing advice" refers to methods and systems that use collected data and generative models to provide appropriate suggestions and information to individual users.

[0507] The embodiment of this invention is based on a system that streamlines communication within the home and promotes information sharing. The server periodically collects and stores the history of electronic data used within the home. Specifically, it aggregates the history obtained from various communication methods used within the home (e.g., messaging apps on smartphones and voice commands via smart speakers) and stores it in a database.

[0508] Next, the server applies natural language processing algorithms to the stored data to analyze each member's communication style, frequently occurring topics, and intentions. Based on the analysis results, it builds a generative AI model tailored to each member. This generative AI model utilizes advanced algorithms such as OpenAI's GPT-3, enabling nuanced communication that takes into account each member's past conversational context.

[0509] Users send questions and requests for information that arise in their daily lives to the server via their smart devices. For example, if a user wants to plan family activities for the next weekend, they might ask their smart speaker for their schedule. The server, upon receiving this question, automatically generates appropriate suggestions and answers based on the AI ​​model of the relevant family member and sends them back to the device. This advice and answers can also be used as part of information sharing for the entire family, and are designed to make it easy for all members to understand the related content.

[0510] For example, if a parent asks a smart speaker, "Tell me some good places for a family picnic next weekend," the system can suggest the best location based on past picnic history and the family's preferences. An example of a prompt used in this case would be, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0511] In this way, the system can improve the overall efficiency of information transfer within the household by facilitating the integration of knowledge and smooth communication, and by providing optimized advice tailored to each individual member.

[0512] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0513] Step 1:

[0514] The server collects the history of electronic data used within the home. Specifically, it periodically retrieves message and voice command history data from communication devices such as smartphones and smart speakers and stores it in a database. The input to this process is raw data from the devices, and the output is an organized history of electronic data.

[0515] Step 2:

[0516] The server applies natural language processing algorithms to the collected electronic data history. In this process, features are extracted from the data, and the communication styles and frequently occurring topics of each household member are analyzed. The input is the organized electronic data history, and the output is the data from which the features have been extracted. Specifically, algorithms such as OpenAI's GPT-3 are used to model the dialogue patterns of each member.

[0517] Step 3:

[0518] The server builds a generative AI model tailored to each household member based on the analyzed features. This generative AI model enables communication simulations that take past conversational context into account. The input is the data from which features have been extracted, and the output is the constructed generative AI model.

[0519] Step 4:

[0520] Users input questions and requests for information to the server via their smart devices. For example, they might tell a smart speaker, "Can you recommend a good picnic spot for next weekend?"

[0521] Step 5:

[0522] The server generates appropriate answers using a generative AI model based on the received request. This process uses historical data and prompts to process information and create relevant information tailored to the user. The input is the user's question, and the output is the generated answer. An example of a prompt is, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0523] Step 6:

[0524] The terminal displays or guides the user with the response received from the server. This allows the user to review the response proposed by the generating AI model and obtain the necessary information. The input for this step is the generated response, and the output is the information provided to the user.

[0525] Step 7:

[0526] Users evaluate the quality of the information and answers provided and send feedback back to the server. This feedback is used to improve the accuracy of the generative AI model. The input is user feedback, and the output is the data that is used to improve accuracy.

[0527] This series of processes effectively optimizes information sharing and communication within the household.

[0528] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0529] This invention is a system that analyzes electronic communication history to generate a digital clone and further combines it with an emotion engine to provide flexible responses based on the user's emotions. Specific embodiments of this invention are described below.

[0530] First, the server periodically retrieves the company's electronic communication history. This includes past message data, sender, recipient, and timestamp. For example, the server collects communication data daily and builds a dataset for each employee.

[0531] Next, the server analyzes the acquired data. Using natural language processing algorithms, it extracts text features and generates a digital clone of each employee based on these features. This digital clone is an individual data model that reflects the communication style and knowledge in the workplace.

[0532] The emotion engine analyzes acquired electronic communication data to identify emotions from messages. For example, the server detects signs of positive, negative, or neutral emotions from specific word choices and expressions.

[0533] When a user submits a question to the server, the server automatically generates a response using a corresponding digital clone. At this point, an emotion engine intervenes, adjusting the response based on the user's current emotional state. For example, if an emotion indicating dissatisfaction or stress is detected, the server will create a response in a more considerate tone.

[0534] The generated response is sent to the terminal and displayed to the user. The user can review the response and provide feedback as needed. This feedback is used by the server to improve the emotion engine and digital clone model.

[0535] For example, if a user asks a question with a nuance like "I'm worried about the recent project," the server will consider the emotion detected by the emotion engine and generate a considerate response such as, "That's a concern. I'll check on the project's progress and suggest ways to improve it."

[0536] This invention enables efficient and emotionally sensitive communication within a company. This, in turn, leads to smoother business operations and improved communication within the organization.

[0537] The following describes the processing flow.

[0538] Step 1:

[0539] The server periodically retrieves communication history via the company's internal electronic communication tools' APIs and stores it in a database. The stored data includes sender, recipient, message text, and timestamp.

[0540] Step 2:

[0541] The server preprocesses the stored data. It formats it into a specific format, performs tokenization and semantic analysis using natural language processing algorithms, and prepares the dataset.

[0542] Step 3:

[0543] The server generates digital clones of each employee based on the prepared dataset. Each clone is an individual model used to generate responses, reflecting the employee's unique communication style and expertise.

[0544] Step 4:

[0545] The emotion engine performs text analysis on electronic communication data to identify emotions such as positive, negative, and neutral. The emotion engine determines emotions based on specific keywords and sentence structure.

[0546] Step 5:

[0547] Users input questions or requests from their terminals and send them to the server. For example, they might ask specific questions such as, "How can I overcome my concerns about a new project?"

[0548] Step 6:

[0549] The server generates a response based on the received question, utilizing the corresponding digital clone. It then adjusts the response to reflect the user's emotions, taking into account the results of the emotion engine's analysis.

[0550] Step 7:

[0551] The server sends back the generated response to the terminal. This response includes an adaptive tone and content that takes emotions into consideration.

[0552] Step 8:

[0553] Users review the responses displayed on their devices and provide feedback on whether the content is appropriate. For example, they might rate the response as helpful.

[0554] Step 9:

[0555] The server adjusts and improves the digital clone model and emotion engine based on user feedback. This maintains improved response accuracy and emotional adaptability for subsequent interactions.

[0556] (Example 2)

[0557] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0558] In today's business environment, communication via electronic means is commonplace. However, conventional systems are insufficient in generating responses that take user emotions into account, hindering effective communication. Furthermore, while providing responses that address emotional needs in internal corporate communication is crucial, the technology to achieve this is lacking. There is a need for highly accurate communication systems that can adjust to user emotions.

[0559] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0560] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, and means for generating individual data models based on the extracted features. This makes it possible to accurately determine the user's emotional state and generate a response adjusted accordingly.

[0561] "Electronic communication history" refers to records of messages, emails, chats, etc., sent and received through a company's internal or external communication network, and includes sender, recipient, timestamp, and message content.

[0562] "Means of acquiring information" refers to the process by which a server collects necessary communication history data from databases and storage on the network.

[0563] "Methods for extracting features" refer to using natural language processing techniques to analyze acquired electronic communication history data and reveal text features such as language usage frequency and specific grammatical patterns.

[0564] An "individualized data model" refers to a digital personality model that reflects each user's communication style and business knowledge, and is a formalized representation of the user's characteristics.

[0565] "Means for automatically generating responses" refers to technologies in which a system automatically creates a response to user input by utilizing a generated data model.

[0566] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of digital clone models and their response generation based on user feedback.

[0567] "Means for determining emotional state" refers to algorithms that analyze text data extracted from communication history to identify emotions such as positive, negative, and neutral.

[0568] "Means of adjusting tone and content" refers to the process of appropriately modifying the wording and content of generated responses to match the emotional state of the identified user.

[0569] This invention is a system that generates a digital clone of a user using their electronic communication history within a company and provides responses that take their emotional state into consideration. An embodiment of this system is shown below.

[0570] First, the server is connected to the company's internal network and periodically retrieves electronic communication history from mail servers and messaging services. This data includes sender, recipient, timestamp, and message content. The server aggregates this data to build a communication history database for each employee.

[0571] Next, this system utilizes natural language processing (NLP) software. The server analyzes the collected communication history data and extracts linguistic features. This extraction process analyzes the grammar, structure, and frequency of terms in the text. Based on this, a digital clone is generated for each employee. Each digital clone is an individual data model that reflects each employee's communication style and work knowledge.

[0572] Furthermore, using an emotion engine, the server determines the user's emotional state from the communication history. It analyzes message text to identify emotions such as positive, negative, or neutral. For example, it infers the user's current emotions from specific words and expressions and adjusts its response accordingly.

[0573] When a user sends a question or request to the server using their device, the server generates a response using an appropriate digital clone. In this process, the server adjusts the response based on information from the emotion engine, ensuring that communication is sensitive to the user's feelings. For example, in response to a user expressing concern about project progress, the server might create a considerate response such as, "Don't worry. We'll check the progress and consider solutions."

[0574] The generated response is sent to the user's device, allowing the user to review the displayed information. Users can also provide feedback on the response, which the server uses to improve the emotion engine and digital clone.

[0575] As a concrete example, a prompt message could be, "Use a digital clone of an employee to generate a thoughtful response based on the user's emotions." This system is expected to improve the quality and efficiency of communication within the company.

[0576] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0577] Step 1:

[0578] The server periodically retrieves electronic communication history from the corporate network. Input includes sender, recipient, timestamp, and message content accessed from mail servers and messaging applications. The server aggregates this data to build a communication history database for each employee. This process results in the output of a digitized communication history for all employees.

[0579] Step 2:

[0580] The server uses natural language processing (NLP) software to analyze the acquired electronic communication history data. The input is the communication history database constructed in Step 1. Using this framework, the server extracts features such as text grammar, terminology, and frequency, and generates digital clones that reflect the user's communication style and business knowledge. The output is the digital clones as individual data models.

[0581] Step 3:

[0582] The server utilizes an emotion engine to determine the user's emotional state from communication history data. The inputs are the digital clone model generated in step 2 and the text data of the electronic communication history. Here, the wording and expressions in the text are analyzed to identify emotions such as positive, negative, and neutral, and the user's emotional state is quantified. The output of this process is the user's current emotional state data.

[0583] Step 4:

[0584] Users can ask questions and make requests to the server using a terminal. The input is a query sent by the user. After receiving this query, the server automatically generates the optimal response, taking into account the digital clone from step 2 and the emotional state from step 3. Specifically, the server adjusts the tone and content of the response based on the user's emotions. The output is the adjusted response message sent to the user.

[0585] Step 5:

[0586] The user reviews the response displayed on the terminal. The user can submit feedback on the response. The input is the user's feedback data. The server receives this feedback and integrates it into the digital clone model and emotion engine, performing an improvement process to enhance the accuracy of future response generation. The output is the improved system behavior.

[0587] (Application Example 2)

[0588] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0589] In today's digital society, electronic communication has become an everyday means of communication. However, inhuman responses and misunderstandings in electronic communication can sometimes degrade the quality of individual communication. Furthermore, within family settings and organizations, there is a need for means to appropriately recognize individual emotions and facilitate smooth relationships. Therefore, a system that utilizes electronic communication history to generate responses adapted to emotions is desired.

[0590] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0591] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, means for generating individual data models based on the extracted features, means for automatically generating responses to new information using the generated data models, means for adjusting the generated responses to take into account the user's emotional state, and means for modifying the data models based on evaluations of the responses. This enables flexible communication that reflects the user's emotions.

[0592] "Electronic communication history" is a general term for the records of emails and messages sent and received by individual users.

[0593] "Means of acquiring information" refers to the function that allows a server to collect necessary data from external databases or networks.

[0594] "Means of analyzing information" refers to the process of analyzing acquired data in detail and extracting important features.

[0595] "Methods for extracting features" refer to techniques that identify useful information from analyzed data and perform analysis based on that information.

[0596] An "individualized data model" is a digital representation that reflects the communication style and emotions of a specific user.

[0597] "Means for automatically generating responses" refers to a method of mechanically creating an appropriate response based on the input information.

[0598] "Means of adjusting based on emotional state" refers to technologies that identify the user's emotional response and modify the content and tone of the response accordingly.

[0599] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of the model and the quality of responses by incorporating feedback from users.

[0600] The system implementing this invention functions as a household robot. A server acquires electronic communication history to form a specific dataset. The acquired data is analyzed using a natural language processing algorithm, which generates individual digital clones. These clones reflect the user's past communication style and emotional tendencies.

[0601] Furthermore, the server uses a sentiment analysis engine to analyze the emotions contained in user inquiries. The sentiment analysis software used here is the "sentiment_analysis" library. Based on the analyzed sentiment data, the tone and content of the response are adjusted. In this process, the "openai" API is utilized to automatically generate appropriate responses.

[0602] The generated response is delivered to the user through a home robot. The user can review the response and provide additional feedback. This feedback is collected on a server and used to improve the accuracy of the digital clone model.

[0603] For example, if a user comments, "Work is tough and I'm tired," a home robot can offer a thoughtful suggestion such as, "How about we watch a relaxing movie together today?" In this case, an example of a prompt for the generative AI model would be:

[0604] "Emotion: Considerate tone. Question: 'Work is tough and I'm tired.' Response:"

[0605] In this way, home robots provide an environment that promotes better communication based on the user's emotions.

[0606] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0607] Step 1:

[0608] The server retrieves electronic communication history from a database. The input is historical data of past emails and messages. Based on this data, it formats information such as date, sender, and recipient to construct a dataset. The output is a processed dataset corresponding to each user.

[0609] Step 2:

[0610] The server uses natural language processing algorithms to analyze the acquired dataset. The input is the dataset formatted in step 1, and this data is subjected to text analysis to extract features. As a result, it outputs digital clones that model each user's communication patterns and emotional tendencies.

[0611] Step 3:

[0612] The terminal (a home robot) uses an emotion analysis engine to receive real-time questions and comments from the user. The input is the user's current inquiry, and emotion analysis detects the emotional tone of its content. The output is the detected emotion data.

[0613] Step 4:

[0614] The server automatically generates responses based on digital clones and emotion data. The input consists of the digital clones from step 2 and the emotion data from step 3, and it generates responses using appropriate prompts that reflect the emotions via the "openai" API. The output is a tailored response that takes the user's emotions into consideration.

[0615] Step 5:

[0616] The device presents the generated response to the user. The input is the response generated in step 4, and interactive communication takes place by conveying this to the user via voice or text. The output is the user's feedback in the form of words or actions.

[0617] Step 6:

[0618] The server collects user feedback and uses it to update the digital clone and response generation algorithms. The input is the user feedback obtained in step 5, and analyzing this feedback and adjusting the model improves the accuracy and appropriateness of the response. The output is the improved digital clone model.

[0619] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0620] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0621] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0622] [Fourth Embodiment]

[0623] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0624] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0625] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0626] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0627] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0628] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0629] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0630] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0631] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0632] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0633] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0634] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0635] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0636] This invention can be implemented as a system for streamlining internal company communication. Specific embodiments are described below.

[0637] The server periodically retrieves and stores the company's electronic communication history. Specifically, after the end of each workday, the server collects message data from all employees via the chat tool's API and stores it in a database. This data includes the message sender, recipient, content, and timestamp.

[0638] Next, the server analyzes the stored data. Using natural language processing algorithms, the server analyzes the message structure and extracts features that allow for interpretation of meaning and intent. This reveals what topics each employee is knowledgeable about and their response style.

[0639] Subsequently, the server generates digital clones of each employee from the analyzed data. These are individualized generative AI models that reflect each employee's communication style and expertise. The digital clones are then adjusted to enable nuanced responses using past conversational context and expertise.

[0640] On the other hand, users can send questions and inquiries that arise in their daily work via a chat tool from their device. For example, if they want to check the progress of a new project, they can send the question to a digital clone of the person in charge.

[0641] Upon receiving this question, the server automatically generates an appropriate answer using the corresponding digital clone. The server then sends the generated answer back to the terminal for display to the user. Because this answer includes knowledge that acts as a liaison between departments, accurate information is provided even for questions concerning the work of other departments.

[0642] Furthermore, users can evaluate the quality of the responses and input feedback into their devices. This feedback is collected by the server and used to further improve the accuracy of the digital clones. This allows the model to continuously improve, and knowledge management within the company is continuously optimized.

[0643] In this way, the present invention is implemented to improve the work efficiency of employees while simultaneously strengthening knowledge sharing throughout the organization.

[0644] The following describes the processing flow.

[0645] Step 1:

[0646] The server retrieves electronic communication history via the company's internal chat tool API at specific times and stores it in a database. This process collects data including the message sender, recipient, content, and timestamp.

[0647] Step 2:

[0648] The server preprocesses the stored chat data. It cleans the text, tokenizes it into a format that is easy for the language model to handle, and removes data noise.

[0649] Step 3:

[0650] The server uses pre-processed data to run natural language processing algorithms and extract key features from the messages. These include frequently occurring terms and each employee's response patterns.

[0651] Step 4:

[0652] The server generates individual digital clone models from the extracted features. These models reflect each employee's expertise and communication style.

[0653] Step 5:

[0654] Users send questions via a chat tool from their device. For example, they might type a question about the progress of a new project.

[0655] Step 6:

[0656] The server receives questions from users and automatically generates answers using the corresponding digital clones.

[0657] Step 7:

[0658] The server sends the generated response to the terminal and displays it to the user. The user can then proceed with their work based on this response.

[0659] Step 8:

[0660] Users verify the accuracy of the answers and enter feedback into their devices. For example, they might rate the answer as "partially correct."

[0661] Step 9:

[0662] The server receives feedback from users and uses it to improve the digital clone model. This allows the model to continue learning and improve its response accuracy.

[0663] (Example 1)

[0664] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0665] In modern organizations, improving the efficiency of internal communication and information sharing are crucial challenges. However, employee knowledge tends to remain tacit and confined to individuals, and important knowledge can be lost due to employee turnover or transfers. Furthermore, to answer questions that arise in daily work quickly and accurately, responses that reflect each employee's work knowledge and communication style are necessary. However, automating this process has been difficult with traditional methods.

[0666] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0667] In this invention, the server includes means for collecting and storing the history of electronic messages, means for analyzing the stored data using natural language processing and extracting features, and means for creating individual generative AI models based on the extracted features. This allows for the construction of individual AI models that reflect each employee's expertise and communication style, enabling the provision of quick and accurate automated responses to user questions. Furthermore, since the AI ​​models can be continuously improved through evaluation of the responses, internal knowledge management and communication efficiency are enhanced.

[0668] "Electronic message history" refers to the record of all messages exchanged through the organization's internal communication channels. This record includes details such as the sender, recipient, content, and timestamp.

[0669] "Natural language processing" refers to the technology used to process human language using computers, and is a means of extracting information from text by performing grammatical analysis and semantic analysis.

[0670] "Features" refer to important attributes and characteristics of information extracted from the content and structure of a message, and these are used to perform more accurate analysis and response generation.

[0671] A "generative AI model" refers to an artificial intelligence model that is trained to automatically analyze information and generate responses for specific tasks.

[0672] A "prompt" refers to an instruction or question that a user inputs to guide a generated AI model.

[0673] "Automated response" refers to a response that is mechanically generated using a generative AI model in response to a user's prompt.

[0674] "Evaluation" refers to the quality assessment conducted by users regarding the accuracy and usefulness of the generated automated responses, and the results of this evaluation are used to improve the AI ​​model.

[0675] This invention can be implemented as a system that streamlines communication within an organization and enables smooth information sharing. This system operates through the collaboration of a server, terminals, and users, with the server playing a central role in performing advanced data analysis and response generation.

[0676] The server first automatically collects communication history from the organization's electronic messaging platform via an API and stores this data in a secure database. This data includes the sender, recipient, content, and timestamp of each message. The server then applies natural language processing algorithms to the stored data to extract important features and patterns from the message text. This analysis is expected to utilize Python natural language processing libraries such as NLTK or spaCy.

[0677] After features are extracted, the server generates generative AI models corresponding to individual employees or members. These generative AI models reflect each employee's communication style and expertise based on their past messaging history. The AI ​​is trained using machine learning libraries such as TensorFlow and PyTorch to generate these models.

[0678] Users can send questions and inquiries related to their daily work to the generated AI model as prompts via their own devices. An example of a prompt a user might send is, "Please tell me the details of the new marketing strategy."

[0679] Based on the prompt received from the user, the server generates an appropriate automated response using the corresponding generative AI model. The generated response is immediately sent back to the terminal and displayed to the user. This process allows the user to obtain information quickly and accurately.

[0680] Furthermore, users can input feedback by evaluating each response. This feedback is collected by the server and used for the continuous improvement of the generated AI model. Through these features, the system will constantly improve and optimize knowledge and information management across the organization. It is also expected that the accuracy of responses to specific prompts will improve, leading to increased employee productivity.

[0681] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0682] Step 1:

[0683] The server periodically collects the history of electronic messages from the organization's electronic messaging platform via an API. The input is message data retrieved from the platform. Specifically, the server runs an automated script daily after the end of the workday to extract message data. This includes the sender, recipient, content, and timestamp for each message. This data is stored in a database for further processing. The output is the structured message data stored in the database.

[0684] Step 2:

[0685] The server analyzes stored message data using natural language processing algorithms. The input is message data from a database. Specifically, the server uses Python's natural language processing library to perform text analysis, including morphological and grammatical analysis. As a result of this analysis, important keywords and phrases are extracted. The output is the extracted feature data.

[0686] Step 3:

[0687] The server creates individual generative AI models based on extracted feature data. The input is the feature data. The server uses machine learning libraries to train generative AI models that reflect each employee's communication style and expertise. This training includes a process of adjusting the model parameters by referring to past messages. The output is the generative AI model for each employee.

[0688] Step 4:

[0689] The user uses their device to send a question as a prompt to the generating AI model. The input is a specific prompt generated by the user. A concrete example would be a question like, "Please tell me the details of the new marketing strategy." The user's input is sent from the device to the server.

[0690] Step 5:

[0691] The server automatically generates a response using the appropriate generative AI model based on the received prompt. The input consists of the prompt and the generative AI model. The server analyzes the prompt and applies it to the AI ​​model to generate an appropriate answer. During this process, the generative AI model constructs an accurate answer by referencing past data. The output is the generated response message.

[0692] Step 6:

[0693] The terminal displays the response received from the server to the user. The input is the response message sent from the server. The terminal formats the received content appropriately and displays it on the user's screen. The output is the response in a format viewable by the user.

[0694] Step 7:

[0695] Users evaluate the quality of the generated responses and enter feedback into the device. This feedback consists of user ratings and comments. Using a feedback form on the device, users provide their opinions on the accuracy and satisfaction level of the responses.

[0696] Step 8:

[0697] The server collects user feedback and uses it to improve the generative AI model. The input is user feedback. The server analyzes the feedback data to help retrain the AI ​​model. This process improves the model's performance and enhances future response accuracy. The output is the improved generative AI model.

[0698] (Application Example 1)

[0699] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0700] Insufficient information sharing among family members can lead to misunderstandings and inefficient communication. Furthermore, it is difficult to accommodate differing communication styles within the family, and there is a lack of means to provide advice tailored to individual needs. It is necessary to address these challenges and achieve smooth communication and information sharing within the family.

[0701] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0702] In this invention, the server includes means for acquiring information from electronic data history, means for analyzing the acquired information and extracting features, means for generating individual generative models based on the extracted features, means for analyzing data obtained from communication means used within the home, and means for providing advice tailored to each member of the household based on the analyzed data. This facilitates information sharing among members of the household and enables effective communication tailored to each member.

[0703] "Electronic data history" refers to a collection of digital information that records past communications and information exchanges within a household or organization.

[0704] "Methods for extracting features" refer to techniques and methods for finding important patterns and trends from acquired data and extracting information that forms the basis of analysis.

[0705] A "generative model" is a digital model that can generate or respond to information based on analyzed features, depending on a specific task or situation.

[0706] "Communication methods" refer to a group of devices and applications used to exchange information in the form of voice, text, video, etc.

[0707] "Means of providing advice" refers to methods and systems that use collected data and generative models to provide appropriate suggestions and information to individual users.

[0708] The embodiment of this invention is based on a system that streamlines communication within the home and promotes information sharing. The server periodically collects and stores the history of electronic data used within the home. Specifically, it aggregates the history obtained from various communication methods used within the home (e.g., messaging apps on smartphones and voice commands via smart speakers) and stores it in a database.

[0709] Next, the server applies natural language processing algorithms to the stored data to analyze each member's communication style, frequently occurring topics, and intentions. Based on the analysis results, it builds a generative AI model tailored to each member. This generative AI model utilizes advanced algorithms such as OpenAI's GPT-3, enabling nuanced communication that takes into account each member's past conversational context.

[0710] Users send questions and requests for information that arise in their daily lives to the server via their smart devices. For example, if a user wants to plan family activities for the next weekend, they might ask their smart speaker for their schedule. The server, upon receiving this question, automatically generates appropriate suggestions and answers based on the AI ​​model of the relevant family member and sends them back to the device. This advice and answers can also be used as part of information sharing for the entire family, and are designed to make it easy for all members to understand the related content.

[0711] For example, if a parent asks a smart speaker, "Tell me some good places for a family picnic next weekend," the system can suggest the best location based on past picnic history and the family's preferences. An example of a prompt used in this case would be, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0712] In this way, the system can improve the overall efficiency of information transfer within the household by facilitating the integration of knowledge and smooth communication, and by providing optimized advice tailored to each individual member.

[0713] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0714] Step 1:

[0715] The server collects the history of electronic data used within the home. Specifically, it periodically retrieves message and voice command history data from communication devices such as smartphones and smart speakers and stores it in a database. The input to this process is raw data from the devices, and the output is an organized history of electronic data.

[0716] Step 2:

[0717] The server applies natural language processing algorithms to the collected electronic data history. In this process, features are extracted from the data, and the communication styles and frequently occurring topics of each household member are analyzed. The input is the organized electronic data history, and the output is the data from which the features have been extracted. Specifically, algorithms such as OpenAI's GPT-3 are used to model the dialogue patterns of each member.

[0718] Step 3:

[0719] The server builds a generative AI model tailored to each household member based on the analyzed features. This generative AI model enables communication simulations that take past conversational context into account. The input is the data from which features have been extracted, and the output is the constructed generative AI model.

[0720] Step 4:

[0721] Users input questions and requests for information to the server via their smart devices. For example, they might tell a smart speaker, "Can you recommend a good picnic spot for next weekend?"

[0722] Step 5:

[0723] The server generates appropriate answers using a generative AI model based on the received request. This process uses historical data and prompts to process information and create relevant information tailored to the user. The input is the user's question, and the output is the generated answer. An example of a prompt is, "Generate a summary of family activity suggestions for a weekend outing based on previous interests and habits."

[0724] Step 6:

[0725] The terminal displays or guides the user with the response received from the server. This allows the user to review the response proposed by the generating AI model and obtain the necessary information. The input for this step is the generated response, and the output is the information provided to the user.

[0726] Step 7:

[0727] Users evaluate the quality of the information and answers provided and send feedback back to the server. This feedback is used to improve the accuracy of the generative AI model. The input is user feedback, and the output is the data that is used to improve accuracy.

[0728] This series of processes effectively optimizes information sharing and communication within the household.

[0729] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0730] This invention is a system that analyzes electronic communication history to generate a digital clone and further combines it with an emotion engine to provide flexible responses based on the user's emotions. Specific embodiments of this invention are described below.

[0731] First, the server periodically retrieves the company's electronic communication history. This includes past message data, sender, recipient, and timestamp. For example, the server collects communication data daily and builds a dataset for each employee.

[0732] Next, the server analyzes the acquired data. Using natural language processing algorithms, it extracts text features and generates a digital clone of each employee based on these features. This digital clone is an individual data model that reflects the communication style and knowledge in the workplace.

[0733] The emotion engine analyzes acquired electronic communication data to identify emotions from messages. For example, the server detects signs of positive, negative, or neutral emotions from specific word choices and expressions.

[0734] When a user submits a question to the server, the server automatically generates a response using a corresponding digital clone. At this point, an emotion engine intervenes, adjusting the response based on the user's current emotional state. For example, if an emotion indicating dissatisfaction or stress is detected, the server will create a response in a more considerate tone.

[0735] The generated response is sent to the terminal and displayed to the user. The user can review the response and provide feedback as needed. This feedback is used by the server to improve the emotion engine and digital clone model.

[0736] For example, if a user asks a question with a nuance like "I'm worried about the recent project," the server will consider the emotion detected by the emotion engine and generate a considerate response such as, "That's a concern. I'll check on the project's progress and suggest ways to improve it."

[0737] This invention enables efficient and emotionally sensitive communication within a company. This, in turn, leads to smoother business operations and improved communication within the organization.

[0738] The following describes the processing flow.

[0739] Step 1:

[0740] The server periodically retrieves communication history via the company's internal electronic communication tools' APIs and stores it in a database. The stored data includes sender, recipient, message text, and timestamp.

[0741] Step 2:

[0742] The server preprocesses the stored data. It formats it into a specific format, performs tokenization and semantic analysis using natural language processing algorithms, and prepares the dataset.

[0743] Step 3:

[0744] The server generates digital clones of each employee based on the prepared dataset. Each clone is an individual model used to generate responses, reflecting the employee's unique communication style and expertise.

[0745] Step 4:

[0746] The emotion engine performs text analysis on electronic communication data to identify emotions such as positive, negative, and neutral. The emotion engine determines emotions based on specific keywords and sentence structure.

[0747] Step 5:

[0748] Users input questions or requests from their terminals and send them to the server. For example, they might ask specific questions such as, "How can I overcome my concerns about a new project?"

[0749] Step 6:

[0750] The server generates a response based on the received question, utilizing the corresponding digital clone. It then adjusts the response to reflect the user's emotions, taking into account the results of the emotion engine's analysis.

[0751] Step 7:

[0752] The server sends back the generated response to the terminal. This response includes an adaptive tone and content that takes emotions into consideration.

[0753] Step 8:

[0754] Users review the responses displayed on their devices and provide feedback on whether the content is appropriate. For example, they might rate the response as helpful.

[0755] Step 9:

[0756] The server adjusts and improves the digital clone model and emotion engine based on user feedback. This maintains improved response accuracy and emotional adaptability for subsequent interactions.

[0757] (Example 2)

[0758] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0759] In today's business environment, communication via electronic means is commonplace. However, conventional systems are insufficient in generating responses that take user emotions into account, hindering effective communication. Furthermore, while providing responses that address emotional needs in internal corporate communication is crucial, the technology to achieve this is lacking. There is a need for highly accurate communication systems that can adjust to user emotions.

[0760] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0761] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, and means for generating individual data models based on the extracted features. This makes it possible to accurately determine the user's emotional state and generate a response adjusted accordingly.

[0762] "Electronic communication history" refers to records of messages, emails, chats, etc., sent and received through a company's internal or external communication network, and includes sender, recipient, timestamp, and message content.

[0763] "Means of acquiring information" refers to the process by which a server collects necessary communication history data from databases and storage on the network.

[0764] "Methods for extracting features" refer to using natural language processing techniques to analyze acquired electronic communication history data and reveal text features such as language usage frequency and specific grammatical patterns.

[0765] An "individualized data model" refers to a digital personality model that reflects each user's communication style and business knowledge, and is a formalized representation of the user's characteristics.

[0766] "Means for automatically generating responses" refers to technologies in which a system automatically creates a response to user input by utilizing a generated data model.

[0767] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of digital clone models and their response generation based on user feedback.

[0768] "Means for determining emotional state" refers to algorithms that analyze text data extracted from communication history to identify emotions such as positive, negative, and neutral.

[0769] "Means of adjusting tone and content" refers to the process of appropriately modifying the wording and content of generated responses to match the emotional state of the identified user.

[0770] This invention is a system that generates a digital clone of a user using their electronic communication history within a company and provides responses that take their emotional state into consideration. An embodiment of this system is shown below.

[0771] First, the server is connected to the company's internal network and periodically retrieves electronic communication history from mail servers and messaging services. This data includes sender, recipient, timestamp, and message content. The server aggregates this data to build a communication history database for each employee.

[0772] Next, this system utilizes natural language processing (NLP) software. The server analyzes the collected communication history data and extracts linguistic features. This extraction process analyzes the grammar, structure, and frequency of terms in the text. Based on this, a digital clone is generated for each employee. Each digital clone is an individual data model that reflects each employee's communication style and work knowledge.

[0773] Furthermore, using an emotion engine, the server determines the user's emotional state from the communication history. It analyzes message text to identify emotions such as positive, negative, or neutral. For example, it infers the user's current emotions from specific words and expressions and adjusts its response accordingly.

[0774] When a user sends a question or request to the server using their device, the server generates a response using an appropriate digital clone. In this process, the server adjusts the response based on information from the emotion engine, ensuring that communication is sensitive to the user's feelings. For example, in response to a user expressing concern about project progress, the server might create a considerate response such as, "Don't worry. We'll check the progress and consider solutions."

[0775] The generated response is sent to the user's device, allowing the user to review the displayed information. Users can also provide feedback on the response, which the server uses to improve the emotion engine and digital clone.

[0776] As a concrete example, a prompt message could be, "Use a digital clone of an employee to generate a thoughtful response based on the user's emotions." This system is expected to improve the quality and efficiency of communication within the company.

[0777] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0778] Step 1:

[0779] The server periodically retrieves electronic communication history from the corporate network. Input includes sender, recipient, timestamp, and message content accessed from mail servers and messaging applications. The server aggregates this data to build a communication history database for each employee. This process results in the output of a digitized communication history for all employees.

[0780] Step 2:

[0781] The server uses natural language processing (NLP) software to analyze the acquired electronic communication history data. The input is the communication history database constructed in Step 1. Using this framework, the server extracts features such as text grammar, terminology, and frequency, and generates digital clones that reflect the user's communication style and business knowledge. The output is the digital clones as individual data models.

[0782] Step 3:

[0783] The server utilizes an emotion engine to determine the user's emotional state from communication history data. The inputs are the digital clone model generated in step 2 and the text data of the electronic communication history. Here, the wording and expressions in the text are analyzed to identify emotions such as positive, negative, and neutral, and the user's emotional state is quantified. The output of this process is the user's current emotional state data.

[0784] Step 4:

[0785] Users can ask questions and make requests to the server using a terminal. The input is a query sent by the user. After receiving this query, the server automatically generates the optimal response, taking into account the digital clone from step 2 and the emotional state from step 3. Specifically, the server adjusts the tone and content of the response based on the user's emotions. The output is the adjusted response message sent to the user.

[0786] Step 5:

[0787] The user reviews the response displayed on the terminal. The user can submit feedback on the response. The input is the user's feedback data. The server receives this feedback and integrates it into the digital clone model and emotion engine, performing an improvement process to enhance the accuracy of future response generation. The output is the improved system behavior.

[0788] (Application Example 2)

[0789] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0790] In today's digital society, electronic communication has become an everyday means of communication. However, inhuman responses and misunderstandings in electronic communication can sometimes degrade the quality of individual communication. Furthermore, within family settings and organizations, there is a need for means to appropriately recognize individual emotions and facilitate smooth relationships. Therefore, a system that utilizes electronic communication history to generate responses adapted to emotions is desired.

[0791] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0792] In this invention, the server includes means for acquiring information from electronic communication history, means for analyzing the acquired information and extracting features, means for generating individual data models based on the extracted features, means for automatically generating responses to new information using the generated data models, means for adjusting the generated responses to take into account the user's emotional state, and means for modifying the data models based on evaluations of the responses. This enables flexible communication that reflects the user's emotions.

[0793] "Electronic communication history" is a general term for the records of emails and messages sent and received by individual users.

[0794] "Means of acquiring information" refers to the function that allows a server to collect necessary data from external databases or networks.

[0795] "Means of analyzing information" refers to the process of analyzing acquired data in detail and extracting important features.

[0796] "Methods for extracting features" refer to techniques that identify useful information from analyzed data and perform analysis based on that information.

[0797] An "individualized data model" is a digital representation that reflects the communication style and emotions of a specific user.

[0798] "Means for automatically generating responses" refers to a method of mechanically creating an appropriate response based on the input information.

[0799] "Means of adjusting based on emotional state" refers to technologies that identify the user's emotional response and modify the content and tone of the response accordingly.

[0800] "Means of modifying data models based on response evaluation" refers to the process of improving the accuracy of the model and the quality of responses by incorporating feedback from users.

[0801] The system implementing this invention functions as a household robot. A server acquires electronic communication history to form a specific dataset. The acquired data is analyzed using a natural language processing algorithm, which generates individual digital clones. These clones reflect the user's past communication style and emotional tendencies.

[0802] Furthermore, the server uses a sentiment analysis engine to analyze the emotions contained in user inquiries. The sentiment analysis software used here is the "sentiment_analysis" library. Based on the analyzed sentiment data, the tone and content of the response are adjusted. In this process, the "openai" API is utilized to automatically generate appropriate responses.

[0803] The generated response is delivered to the user through a home robot. The user can review the response and provide additional feedback. This feedback is collected on a server and used to improve the accuracy of the digital clone model.

[0804] For example, if a user comments, "Work is tough and I'm tired," a home robot can offer a thoughtful suggestion such as, "How about we watch a relaxing movie together today?" In this case, an example of a prompt for the generative AI model would be:

[0805] "Emotion: Considerate tone. Question: 'Work is tough and I'm tired.' Response:"

[0806] In this way, home robots provide an environment that promotes better communication based on the user's emotions.

[0807] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0808] Step 1:

[0809] The server retrieves electronic communication history from a database. The input is historical data of past emails and messages. Based on this data, it formats information such as date, sender, and recipient to construct a dataset. The output is a processed dataset corresponding to each user.

[0810] Step 2:

[0811] The server uses natural language processing algorithms to analyze the acquired dataset. The input is the dataset formatted in step 1, and this data is subjected to text analysis to extract features. As a result, it outputs digital clones that model each user's communication patterns and emotional tendencies.

[0812] Step 3:

[0813] The terminal (a home robot) uses an emotion analysis engine to receive real-time questions and comments from the user. The input is the user's current inquiry, and emotion analysis detects the emotional tone of its content. The output is the detected emotion data.

[0814] Step 4:

[0815] The server automatically generates responses based on digital clones and emotion data. The input consists of the digital clones from step 2 and the emotion data from step 3, and it generates responses using appropriate prompts that reflect the emotions via the "openai" API. The output is a tailored response that takes the user's emotions into consideration.

[0816] Step 5:

[0817] The device presents the generated response to the user. The input is the response generated in step 4, and interactive communication takes place by conveying this to the user via voice or text. The output is the user's feedback in the form of words or actions.

[0818] Step 6:

[0819] The server collects user feedback and uses it to update the digital clone and response generation algorithms. The input is the user feedback obtained in step 5, and analyzing this feedback and adjusting the model improves the accuracy and appropriateness of the response. The output is the improved digital clone model.

[0820] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0821] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0822] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0823] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0824] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0825] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0826] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0827] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0828] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0829] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0830] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0831] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0832] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0833] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0834] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0835] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0836] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0837] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0838] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0839] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0840] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0841] The following is further disclosed regarding the embodiments described above.

[0842] (Claim 1)

[0843] Means of obtaining information from electronic communication history,

[0844] A means of analyzing acquired information and extracting features,

[0845] A means for generating individual data models based on extracted features,

[0846] A means for automatically generating responses to new information using the generated data model,

[0847] A means for modifying the data model based on the evaluation of the aforementioned response,

[0848] A system that includes this.

[0849] (Claim 2)

[0850] The system according to claim 1, which applies a natural language processing algorithm in the generation of individual data models.

[0851] (Claim 3)

[0852] The system according to claim 1, which retains information from the electronic communication history of former employees and prevents the loss of knowledge within the organization.

[0853] "Example 1"

[0854] (Claim 1)

[0855] A means for collecting and storing the history of electronic messages,

[0856] A method for analyzing stored data using natural language processing and extracting features,

[0857] A means for creating individual generative AI models based on extracted features,

[0858] A means of generating an automated response to a user prompt using the generated AI model,

[0859] A means of collecting user evaluations of the generated responses and improving the AI ​​model,

[0860] A system that includes this.

[0861] (Claim 2)

[0862] The system according to claim 1, which reflects past dialogue context in individual generative AI models.

[0863] (Claim 3)

[0864] The system according to claim 1, which reflects the expertise and communication style of each member of the organization into individual AI models.

[0865] "Application Example 1"

[0866] (Claim 1)

[0867] Means of obtaining information from electronic data history,

[0868] A means of analyzing acquired information and extracting features,

[0869] A means for generating individual generative models based on extracted features,

[0870] A means of automatically generating answers to new information using the generated generative model,

[0871] A means of analyzing data obtained from communication methods used within the home,

[0872] A means of providing personalized advice to family members based on the analyzed data,

[0873] A means of adjusting the generative model based on the evaluation of the above response,

[0874] A system that includes this.

[0875] (Claim 2)

[0876] The system according to claim 1, which applies a natural language processing technique in the generation of individual generative models.

[0877] (Claim 3)

[0878] The system according to claim 1, which retains information from in-home communication data and facilitates information sharing within the home.

[0879] "Example 2 of combining an emotion engine"

[0880] (Claim 1)

[0881] Means of obtaining information from electronic communication history,

[0882] A means of analyzing acquired information and extracting features,

[0883] A means for generating individual data models based on extracted features,

[0884] A means for automatically generating responses to new information using the generated data model,

[0885] A means for modifying the data model based on the evaluation of the aforementioned response,

[0886] A means of determining the user's emotional state based on the extracted information,

[0887] Means of adjusting the tone and content of the response in consideration of the judged emotional state,

[0888] A system that includes this.

[0889] (Claim 2)

[0890] The system according to claim 1, which applies a natural language processing algorithm in the generation of individual data models.

[0891] (Claim 3)

[0892] The system according to claim 1, which retains information from the electronic communication history of former employees and prevents the loss of knowledge within the organization.

[0893] "Application example 2 when combining with an emotional engine"

[0894] (Claim 1)

[0895] Means of obtaining information from electronic communication history,

[0896] A means of analyzing acquired information and extracting features,

[0897] A means for generating individual data models based on extracted features,

[0898] A means for automatically generating responses to new information using the generated data model,

[0899] A means by which the generated response is adjusted taking into account the user's emotional state,

[0900] A means for modifying the data model based on the evaluation of the aforementioned response,

[0901] A system that includes this.

[0902] (Claim 2)

[0903] The system according to claim 1, which applies a natural language processing algorithm to the generation of individual data models and proposes a response that is in line with the user's emotional state.

[0904] (Claim 3)

[0905] The system according to claim 1, which facilitates communication between users by considering the electronic communication history and the emotional state of the users in a home environment. [Explanation of symbols]

[0906] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means of obtaining information from electronic data history, A means of analyzing acquired information and extracting features, A means for generating individual generative models based on extracted features, A means of automatically generating answers to new information using the generated generative model, A means of analyzing data obtained from communication methods used within the home, A means of providing personalized advice to family members based on the analyzed data, A means of adjusting the generative model based on the evaluation of the above response, A system that includes this.

2. The system according to claim 1, wherein a natural language processing technique is applied in the generation of individual generative models.

3. The system according to claim 1, which retains information from in-home communication data and facilitates information sharing within the home.

Images