system

The system addresses the challenge of obtaining reliable advice by incorporating a terminal, server, and generative AI model to provide personalized and credible advice based on celebrity insights, improving life choice decisions.

JP2026035122APending Publication Date: 2026-03-04SOFTBANK GROUP CORP
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2024137965
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2026-03-04

AI Technical Summary

Technical Problem

Individuals face challenges in obtaining reliable advice for life choices due to limited opportunities for direct advice from celebrities and questionable credibility of existing generative AI models.

Method used

A system that includes a terminal for user input, a server storing celebrity quotes and experiences, a generative AI model trained on a curated dataset, and internal algorithms for advice verification, ensuring advice appropriateness and reliability.

Benefits of technology

Provides users with highly reliable and specific advice tailored to their concerns, enhancing decision-making in life choices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026035122000001_ABST
    Figure 2026035122000001_ABST
Patent Text Reader

Abstract

Provide a system. [Solution] a terminal that receives a request from a user; A server that stores famous quotes, personal experiences, and past advice as a dataset, and A server that analyzes the received requests and selects and trains a generative AI model based on the required training dataset; a server that generates advice in response to a user request using the selected generative AI model; a terminal for providing the generated advice to a user; A system including:
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

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

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

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

[0004] In modern society, many people are faced with life choices and seek reliable advice. However, opportunities to obtain direct advice from celebrities are limited, and the credibility and appropriateness of advice provided by existing generative AI models remain questionable. The present invention aims to solve these problems and provide a system that allows users to easily obtain reliable advice when making life decisions. [Means for solving the problem]

[0005] The present invention is a system that includes a terminal that receives requests from users, a server that stores famous quotes, personal experiences, and past advice from celebrities as a dataset, a server that analyzes the received requests and selects and trains a generative AI model based on the necessary training dataset, a server that generates advice in response to the user's request using the selected generative AI model, and a terminal that provides the generated advice to the user. Furthermore, by formulating the dataset with the celebrity's permission and supervision and providing a means for generating reliable advice, users can receive advice with peace of mind. Furthermore, by providing a means for verifying the appropriateness of the generated advice using an internal algorithm and adjusting it as necessary, the reliability of the advice can be further improved.

[0006] A "user" is an individual or group who uses this system to obtain advice about their life path.

[0007] "Terminal" refers to a device that allows a user to access the system, input requests, and receive advice. This includes personal computers, smartphones, tablets, etc.

[0008] A "request" is input data that describes a specific question or situation that a user is struggling with regarding life choices.

[0009] "Server" means a central computer system for storing and analyzing data and for training and running generative AI models.

[0010] A "dataset" refers to a collection, organization, and storage of famous quotes, personal experiences, and past advice from famous people as written information.

[0011] A "generative AI model" is an algorithmic model of artificial intelligence trained on a dataset of celebrities and capable of generating advice based on user requests.

[0012] "Advice" is the specific advice or recommendation that a generative AI model provides in response to a user request.

[0013] "Internal algorithms" are programmatic methods and procedures for verifying the appropriateness and reliability of the advice generated and adjusting it as necessary.

[0014] "Permission" refers to formal consent given by celebrities and relevant rights holders to use the dataset.

[0015] "Supervision" refers to the act of having celebrities and experts verify the accuracy and reliability of a dataset and review its contents to ensure that appropriate advice is generated. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] The present invention is a system that enables users to obtain reliable advice to solve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores famous quotes, personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models.

[0038] Program processing

[0039] Data Collection and Training

[0040] The server collects quotes, personal experiences, and past advice from famous people online and in books. The collected data is saved in text format and then edited by experts to be ready for use as a training dataset. This dataset is used to train a generative AI model. The trained model reflects the thinking patterns and advice of famous people.

[0041] Receiving and parsing the request

[0042] The user accesses the system and inputs specific concerns or questions into the terminal. For example, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[0043] AI model selection and generation

[0044] The server analyzes the received request and selects the celebrity generative AI model that best suits the request. The selection criteria are the relevance between the request content and the training data. The selected generative AI model is used to generate advice that meets the user's request. The generation process ensures that reliable advice is provided based on the celebrity's thought patterns and past advice that the model has learned.

[0045] Providing advice

[0046] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is not overly biased and whether it is expressed in an appropriate context. If the verified advice is deemed appropriate, it is sent to the terminal and displayed to the user. The user can then consider their own choices and make a decision based on this advice.

[0047] Specific examples

[0048] Example 1: Career Choices

[0049] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0050] 2. The device receives this request and sends it to the server.

[0051] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model with extensive experience in career decisions.

[0052] 4. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks."

[0053] 5. The server sends the verified advice to the terminal and displays it to the user.

[0054] Example 2: Academic direction

[0055] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0056] 2. The device receives this request and sends it to the server.

[0057] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model that is good at giving academic advice.

[0058] 4. Based on the AI ​​model, the server generates advice such as, "Value your own interests and curiosity, and make a choice based on which one is best suited to you."

[0059] 5. The server sends the verified advice to the terminal and displays it to the user.

[0060] This system allows users to obtain reliable advice that can be used effectively in making life choices.

[0061] The processing flow will be explained below.

[0062] Step 1:

[0063] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[0064] Step 2:

[0065] The server uses the training dataset to train the generative AI model. Specifically, it uses natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[0066] Step 3:

[0067] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[0068] Step 4:

[0069] The terminal receives a request input by the user and transmits the content in text format to the server, which then formats the request appropriately for the system.

[0070] Step 5:

[0071] The server analyzes the received request. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable AI model to generate a celebrity.

[0072] Step 6:

[0073] The server uses the selected generative AI model to generate advice for the user's request, referencing the thinking patterns and past advice of famous people that the model has learned, and generates advice that is specifically suited to the request.

[0074] Step 7:

[0075] The server uses an internal algorithm to verify the appropriateness of the advice it generates, for example, by checking that the advice is not overly biased and that it uses appropriate language in the context. If necessary, it adjusts the advice it generates.

[0076] Step 8:

[0077] The server then sends the final verified and adjusted advice to the terminal, where it is displayed to the user.

[0078] Step 9:

[0079] Users receive advice displayed on their devices and use it to consider their choices, helping them make better decisions about their life paths.

[0080] In this way, the system provides users with reliable advice and allows them to obtain useful information in making life choices.

[0081] Example 1

[0082] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0083] In today's world, when individuals face various life choices, it is difficult to obtain advice based on reliable information. Furthermore, existing advice systems are general and cannot provide specific advice tailored to individual concerns or questions. Therefore, there is a need for the development of a system that can address the specific problems faced by individual users and provide accurate advice based on deep insight and experience.

[0084] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0085] In this invention, the server includes an information processing device that receives requests from users, a data storage device that uses a database to store quotes and personal experiences, an analysis device that analyzes the received requests using a natural language processing tool, a learning device that selects and trains an optimal generative AI model based on the analysis results, a generation device that generates advice in response to the user request using the selected generative AI model, and a display device that provides the generated advice to the user. This makes it possible to provide highly reliable and specific advice based on the concerns and questions of individual users.

[0086] An "information processing device" is a device capable of receiving input from a user and forwarding it to other system components.

[0087] A "data storage device" is a device that stores data such as famous quotes and personal experiences in a database and has the function of retrieving that data as needed.

[0088] An "analysis device" is a device that uses natural language processing tools to analyze a received request and has the function of understanding the content of the request.

[0089] A "learning device" is a device that has the function of selecting the optimal generative AI model based on the analysis results and training that model.

[0090] A "generation device" is a device that has the function of generating specific advice in response to a user request using a selected generative AI model.

[0091] A "display device" is a device that has the function of providing and displaying the generated advice to the user.

[0092] A "natural language processing tool" is software or a library that analyzes received text data and understands and classifies its content.

[0093] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task based on a training dataset.

[0094] The present invention is a system that allows users to get reliable advice to solve specific questions and concerns about their life path. The system includes the following main components:

[0095] Data collection

[0096] The server collects quotes, personal experiences, and past advice from famous people from the internet and books. For this collection, it uses web scraping tools such as Scrapy and Beautiful Soup. The collected data is then stored in a database in text format.

[0097] Data Preparation and Training

[0098] The server then uses Hugging Face's Transformers library to organize the collected text data into a training dataset, allowing the generative AI model to learn the thought patterns and advice of celebrities. This organization process also uses natural language processing libraries such as NLTK (Natural Language Toolkit) and spaCy.

[0099] Receiving a user request

[0100] Users access the system using a dedicated app or a web browser and enter their specific concerns or questions. The text is then sent to the server via web application frameworks such as Flask and Django.

[0101] Parsing the request

[0102] The server uses spaCy to perform natural language analysis on the received request. Based on the analysis results, the request content is classified and an appropriate generative AI model is selected. The selection criterion in this case is the relevance of the request content to the training dataset.

[0103] Generating Advice

[0104] The server uses the selected generative AI model to generate specific advice based on the user's request. The generation process incorporates the thought patterns and past advice of famous people that have been trained in advance.

[0105] Validating and providing advice

[0106] The generated advice is then validated using an internal algorithm to ensure it is appropriate. Specifically, it checks whether the advice is overly biased and whether it is expressed appropriately for the context. After this validation is complete, the server sends the advice to the terminal via Flask or Django and displays it to the user.

[0107] Specific examples

[0108] Example 1: Career Choices

[0109] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0110] 2. The device receives this request and sends it to the server via Flask.

[0111] 3. The server uses spaCy to analyze the request and extract keywords such as "job change" and "workplace."

[0112] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as "Pursue your passion and take risks."

[0113] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[0114] Example 2: Academic direction

[0115] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0116] 2. The device receives this request and sends it to the server via Flask.

[0117] 3. The server uses spaCy to analyze the request and extract keywords such as "science" and "humanities."

[0118] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as, "Value your interests and curiosity, and make a choice based on which one is most suited to you."

[0119] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[0120] As described above, this system allows users to obtain reliable and specific advice, and the entire system operates through a consistent process, providing a fast and appropriate response to user requests.

[0121] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0122] Step 1:

[0123] Users access the system using a dedicated app or a web browser and enter specific concerns or questions. The entered text is sent to the server via Flask or Django.

[0124] Specific behavior: The user enters the following into the web form: "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job," and clicks the submit button. This is the input.

[0125] Step 2:

[0126] The terminal receives this request and sends it to the server as text data. The server receives the request.

[0127] Specific operation: The terminal receives input and sends a POST request to the server via Flask or Django in JSON or text format, which is the output to the server.

[0128] Step 3:

[0129] The request received by the server is analyzed. spaCy, a natural language processing tool, is used to understand the content of the request and extract keywords. This is the input. Based on the analysis results, the content of the request is classified.

[0130] Specific operation: The server analyzes a request such as "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job" using spaCy, and extracts keywords such as "job change" and "workplace." This is the output of the analysis.

[0131] Step 4:

[0132] The server selects and trains the optimal generative AI model based on the analysis results. The selection criteria are the relevance of the request content to the training dataset, which is the input. A learning device is used to select the model and conduct additional training if necessary.

[0133] Specific operation: The server selects a generative AI model for "Career Advisor" based on the analyzed keywords, and trains the model for the number of epochs using the training dataset. This is the output of model selection and training.

[0134] Step 5:

[0135] The server uses the selected generative AI model to generate specific advice in response to the user's request. The input is the user's request and the selected AI model. The generation process reflects the thinking patterns and past advice of famous people that have been trained in advance.

[0136] How it works: The server inputs a prompt such as "I'm thinking about changing jobs, but should I try a new job?" into the career advisor's generative AI model, and gets the generated advice "Follow your passion and take risks." This is the output of the generation.

[0137] Step 6:

[0138] The server validates the appropriateness of the generated advice using an internal algorithm. The input is the generated advice. It checks whether the advice is overly biased and whether it is expressed appropriately for the context.

[0139] What happens: The server validates the generated advice, "Follow your passion and take risks," to check for inappropriate content. This is the validation output:

[0140] Step 7:

[0141] If the server determines that the advice is valid, it sends it to the terminal via Flask or Django and displays it to the user. The input is the validated advice.

[0142] What happens: The server sends the advice "Follow your passion and take risks" to the terminal through the Flask and Django application and displays it on the user's screen. This is the display output:

[0143] (Application example 1)

[0144] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0145] In traditional brick-and-mortar stores, it was difficult for customers to ask questions about specific products or services in real time and receive reliable advice. There were also few ways to effectively utilize advice based on the words and personal experiences of celebrities. As a result, customers lacked the information they needed to make the best choice for themselves, leading to problems such as reduced purchasing efficiency and satisfaction.

[0146] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0147] In this invention, the server includes a means for receiving requests from users, a means for storing famous people's quotes, personal experiences, and past advice as a dataset, and a means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset. This allows customers to receive product advice in real time in a physical store. Furthermore, the selected generative AI model can be used to provide reliable advice, quickly and accurately providing customers with information to help them make the best choice for themselves.

[0148] A "terminal that receives requests from a user" is a device through which a user inputs questions or requests and transmits that information to other system components.

[0149] An "information processing device that stores the words, experiences, and past advice of famous people as a data set" is a device that collects, organizes, and stores data such as the statements and experiences of famous people.

[0150] An "information processing device that analyzes received requests and selects and trains a generative AI model based on the required training data set" is a device that analyzes requests from users, selects an AI model based on appropriate training data, and trains it.

[0151] An "information processing device that generates advice in response to a user request using a selected generative AI model" is a device that generates specific advice in response to a user question using a selected AI model.

[0152] "Means that enable customers to obtain product advice using a terminal in a physical store" refers to a system that enables customers to obtain advice on product selection and purchases through a terminal in a physical store environment.

[0153] The "terminal for providing generated advice to a user" is a device for displaying, notifying, or communicating generated advice to a user.

[0154] "Formalizing the dataset with the permission and supervision of the celebrity" means recording the statements and activities of the celebrity as official data with the celebrity's permission and supervision.

[0155] A "means for generating reliable advice" is a process or mechanism for providing accurate and useful advice to users based on reliable sources of information and data.

[0156] "Verifying the appropriateness of the generated advice using an internal algorithm" means confirming the accuracy and appropriateness of the advice generated using an internal algorithm.

[0157] "Means for adjustment as necessary" refers to a mechanism for correcting or improving the content of advice generated if it is inappropriate.

[0158] This invention is a system that allows customers to ask questions about product selection and purchases in a physical store and receive reliable advice in real time. This system consists of a terminal where the user inputs their questions, an information processing device (server) that stores and utilizes quotes and personal experiences of celebrities and past advice as a dataset, and an information processing device that trains and executes a generative AI model.

[0159] Hardware and Software

[0160] Hardware:

[0161] Smartphones, smart glasses (to input user questions and display advice)

[0162] Servers (for storing data, training and running AI models)

[0163] software:

[0164] OpenAI® API (AI model creation and execution)

[0165] HTTP request library (requests)

[0166] Program processing overview

[0167] The server collects the words, experiences, and past advice of celebrities from online and offline sources and stores them as a dataset. The collected data is then used to train a generative AI model. This model learns the thought patterns and advice of celebrities and generates advice accordingly.

[0168] When a user enters a question using a smartphone or smart glasses in a physical store, the device sends the request in text format to the server. The server analyzes the request, selects the most appropriate generative AI model, and generates advice. The generated advice is verified for appropriateness by an internal algorithm, and if deemed appropriate, it is sent to the device and displayed to the user.

[0169] Specific examples

[0170] Selecting products in the store

[0171] A customer in a physical store puts on smart glasses and inputs a question such as, "Which brand is best when choosing new sneakers?"

[0172] Example prompt sentence:

[0173] Customer Question: What brand is best when choosing new sneakers?

[0174] You are an AI that provides advice based on the experiences and words of famous athletes. Please advise users on which brand of sneakers to choose by referring to the following dataset:

[0175] Dataset:

[0176] 1. Comments from famous athletes about their sneaker choices

[0177] 2. Expert-recommended brands

[0178] 3. Industry magazine rankings

[0179] This system allows customers to instantly receive helpful advice on product selection in-store and receive assistance in making the best choices.

[0180] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0181] Step 1:

[0182] The server collects quotes, personal experiences, and past advice from famous people both online and offline. The collected data is stored in text format and then compiled into training data through expert supervision. This allows the server to maintain a highly reliable data set.

[0183] Input: Text data collected from the internet and books

[0184] Output: A curated training dataset curated by experts

[0185] Specific operations: Data is obtained from online databases and books using web scraping and OCR technology, and then reviewed and tagged by experts.

[0186] Step 2:

[0187] The server trains the generative AI model using a curated training dataset, which learns the thought patterns and advice of celebrities.

[0188] Input: Training dataset

[0189] Output: A trained generative AI model

[0190] Specific operations: Using the training dataset, train the model using an AI model training framework (e.g., TENSORFLOW (registered trademark), PyTorch).

[0191] Step 3:

[0192] Users use their smartphones or smart glasses in a physical store to input questions into the terminal, which then sends the questions in text format to the server.

[0193] Input: User question text

[0194] Output: Question text sent to the server

[0195] Specific operation: The user enters a question into the application's interface, which is then sent by the device to the server.

[0196] Step 4:

[0197] The server analyzes the received request and selects the most suitable generative AI model based on the relevance of the request content to the training data.

[0198] Input: User question text

[0199] Output: The selected generative AI model

[0200] Specific operation: The server uses natural language processing technology to analyze the question text and select the corresponding generative AI model.

[0201] Step 5:

[0202] The server uses the selected generative AI model to generate advice for the user's question. The generated advice is based on the thought patterns and past advice of famous people that the model has learned.

[0203] Input: User question text, generative AI model

[0204] Output: The generated advice

[0205] Specific operation: The server inputs the prompt sentence into the selected generative AI model and performs text generation using AI.

[0206] Step 6:

[0207] The server uses an internal algorithm to verify the appropriateness of the advice it generates, ensuring that it is expressed in the right context and is not overly biased.

[0208] Input: Generated advice

[0209] Output: Verified advice

[0210] What happens: The server uses internal algorithms and rule-based validation systems to check the content of the advice generated.

[0211] Step 7:

[0212] The server sends the verified advice to the user's device, which displays it to the user, who then considers his or her options and makes a decision based on the advice displayed.

[0213] Input: Verified Advice

[0214] Output: Advice displayed by the user

[0215] Specific operation: The server sends the advice in an appropriate format to the terminal, and the terminal displays the advice on the screen.

[0216] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0217] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components include a device that receives user requests, a server that stores famous quotes and personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models. The system also incorporates an emotion engine that recognizes the user's emotions and can analyze the user's emotional state when the request is made and tailor the advice accordingly.

[0218] Program processing

[0219] Data Collection and Training

[0220] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and edited by experts to create a training dataset. A generative AI model is trained using this dataset. The trained model reflects the thinking patterns and advice of famous people.

[0221] Receiving requests and analyzing sentiment

[0222] The user accesses the system and inputs specific concerns or questions in text format from a terminal. For example, they might input something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[0223] Analysis by emotion engine

[0224] The device incorporates an emotion engine that recognizes the user's emotional state when inputting a request. The emotion engine analyzes the user's emotions through analysis of the user's voice input and text, and sends the analysis results to the server. For example, if the user is feeling anxious, that emotional state will be included.

[0225] Request and emotional state analysis

[0226] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model and generates advice taking into account the user's emotional state.

[0227] AI model selection and generation

[0228] The server uses the selected generative AI model to generate advice based on the user's request and emotional state, and provides advice that matches the user's emotional state by referencing the thought patterns and past advice of famous people that the model has learned.

[0229] Validating and providing advice

[0230] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is overly biased and whether it uses appropriate language in line with the context. If necessary, it adjusts the generated advice. If the final verified and adjusted advice is deemed appropriate, it is sent to the device and displayed to the user. The user can then consider their own choices based on this advice.

[0231] Specific examples

[0232] Example 1: Career Choice and Sentiment Analysis

[0233] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0234] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[0235] 3. The device sends the request and emotional state to the server.

[0236] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[0237] 5. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks," and adds phrases to ease anxiety.

[0238] 6. The server sends the verified advice to the terminal and displays it to the user.

[0239] Example 2: Academic orientation and sentiment analysis

[0240] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0241] 2. The device receives this request and uses its emotion engine to analyze that the user is excited.

[0242] 3. The device sends the request and emotional state to the server.

[0243] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[0244] 5. Based on the AI ​​model, the server generates advice such as "Value your interests and curiosity, and make a choice based on which direction they are directed," and adds expressions to help the user stay calm.

[0245] 6. The server sends the verified advice to the terminal and displays it to the user.

[0246] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[0247] The processing flow will be explained below.

[0248] Step 1:

[0249] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[0250] Step 2:

[0251] The server uses the training dataset to train the generative AI model, specifically using natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[0252] Step 3:

[0253] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[0254] Step 4:

[0255] The device receives requests entered by the user and sends them in text format to the server. The device's built-in emotion engine analyzes the user's emotional state as they speak, extracting emotional states such as joy, sadness, anxiety, and excitement based on the user's voice input and text information.

[0256] Step 5:

[0257] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model. The selection criteria are the relevance of the request content to the training data and the user's emotional state.

[0258] Step 6:

[0259] The server uses the selected generative AI model to generate advice in response to the user's request. The generated advice is based on the thought patterns and past advice of famous people that the model has learned. In addition, the expression and content are adjusted according to the user's emotional state. For example, if a user is feeling anxious, reassuring expressions are used.

[0260] Step 7:

[0261] The server uses an internal algorithm to verify the appropriateness of the generated advice. It checks whether the advice is excessively biased and whether appropriate expressions are used for the context. If necessary, it makes corrections to the generated advice.

[0262] Step 8:

[0263] The server then sends the final verified and corrected advice to the terminal, which then displays it to the user.

[0264] Step 9:

[0265] Users receive advice displayed on their devices and use it to consider their own choices. For example, a user receiving advice about changing jobs can use the advice to determine the direction they truly want to take. Furthermore, advice appropriately adjusted by the emotion engine is tailored to the user's emotional state, allowing them to accept the advice with greater confidence.

[0266] In this way, the system provides users with reliable advice that takes into account their emotional state, enabling them to obtain useful information for making life choices.

[0267] Example 2

[0268] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0269] In modern society, many people are concerned about their careers, studies, and life paths. However, it is not easy to get reliable advice. In addition, there are limited advice-giving systems that take the user's emotional state into account, and general advice often lacks emotional consideration. This makes it difficult for users to make appropriate and meaningful decisions.

[0270] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a terminal that receives a request from a user, a means for saving famous quotes, personal experiences, and past advice as a dataset, a means for analyzing the received request and extracting necessary keywords and key points using natural language processing technology, a sentiment analysis means that recognizes the user's emotional state and adjusts the content of the advice based on the analysis results, a means for selecting and training a generative AI model based on a necessary training dataset, a means for generating advice in response to the user's request using the selected generative AI model, a means for verifying the appropriateness of the generated advice using an internal algorithm and adjusting it as necessary, and a terminal that provides the generated advice to the user. This makes it possible to provide highly reliable advice that takes the user's emotional state into consideration.

[0271] "User" refers to a person who accesses the system, enters their concerns and questions, and seeks advice.

[0272] "Terminal" means a device that allows a user to access the system and input / submit requests. Examples include PCs and smartphones.

[0273] A "quote" is a commonly quoted phrase or expression uttered by a famous or influential individual.

[0274] "Testimonials" refer to stories or accounts recorded by individuals about specific events or experiences.

[0275] "Past advice" refers to advice or suggestions given to others in the past.

[0276] "Dataset" refers to a set of data collected and organized for use in the System.

[0277] A "server" is a computer system that receives requests from users, processes them, and generates and provides advice.

[0278] "Natural language processing technology" refers to the techniques and algorithms that enable computers to understand and process human language.

[0279] "Keywords" refer to words or phrases that play a significant role in a user's request or text.

[0280] "Emotional state" refers to the emotions or psychological state a user experiences in a particular situation.

[0281] "Emotion analysis means" refers to the technology or algorithm that recognizes emotions from user requests or voice and provides the analysis results.

[0282] A "generative AI model" is an artificial intelligence model that is trained using a large dataset and is used to generate advice in response to a user request.

[0283] "Relevance" refers to the unbiased nature of the advice generated, and whether it is natural, contextual, and appropriate.

[0284] "Internal algorithm" refers to the calculation methods and programs used within the system.

[0285] "Validation" refers to the process of evaluating whether the generated advice is appropriate.

[0286] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores quotes, personal experiences, and past advice as a dataset, and a server that trains and executes a generative AI model. The system also incorporates an emotion analysis means that recognizes the user's emotions, and can analyze the user's emotional state when the request is made and adjust the content of the advice accordingly.

[0287] The server uses Python scripts and web analytics tools to collect quotes and testimonials from online sources and books, and store them in a database in text format. Specifically, the Beautiful Soup and Requests libraries are used to scrape data from websites, the Pandas library is used to format and clean the data, and then the data is stored in a MySQL® database via SQLAlchemy.

[0288] The collected dataset is used to train a generative AI model (e.g., GPT-3®) using GPUs on cloud computing platforms (e.g., AWS® or Google® Cloud). The training process is carried out over hundreds of epochs using libraries such as TensorFlow and PyTorch.

[0289] Users access the web application using a device such as a PC or smartphone and enter specific concerns or questions in text format. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The device then asynchronously sends this request to the server using JavaScript (registered trademark) (AJAX).

[0290] The device uses emotion analysis to recognize the user's emotional state from the input text. For example, it uses the BERT model with the Hugging Face transformers library to classify emotions and obtain emotion labels such as "anxiety" or "excitement." The analysis results are sent to the server in JSON format.

[0291] The server combines the request and the emotional state and uses natural language processing (NLP) techniques to analyze the request. For example, it uses the spaCy library to tokenize the text, extract nouns and verbs, and identify keywords. Keywords such as "job change," "new workplace," and "challenge" are extracted.

[0292] The server selects the most suitable generative AI model based on the analysis results. It uses a keyword matching algorithm to select the best one from multiple celebrity models. If the keyword "challenge" is included, it selects a model that provides adventurous advice.

[0293] Using the selected generative AI model, the server generates advice based on the user's request and emotional state. For example, using GPT-3, the server generates advice such as, "Identify your passion and have the courage to take risks." The prompt used is, "I'm thinking about changing jobs. I'm feeling anxious. What should I do?"

[0294] The server uses internal algorithms to verify the appropriateness of the generated advice and adjusts it as needed, for example by using rule-based filters to ensure that it does not contain extreme or inappropriate language. Finally, advice that passes the verification is sent to the device and displayed to the user.

[0295] Specific examples of prompts are as follows:

[0296] "I'm thinking about changing jobs, but I'm not sure whether I should stay where I am or try something new. What should I do? I'm feeling anxious."

[0297] "I'm worried about my further education. I'm not sure whether to go into science or humanities, but I'm feeling excited. I'd appreciate some advice."

[0298] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[0299] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0300] Step 1: User request input

[0301] A user enters specific questions or concerns about their life path into a text box in a web application. For example, they might enter, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job." This is the input data. The device asynchronously sends the text entered by the user to the server using AJAX. This sent data is the input for the next step.

[0302] Step 2: Receiving the request

[0303] The terminal receives text input from the user and passes it to the emotion analysis means. The input data is the text information sent by the user. The server receives the text received via an AJAX request for analysis. The output is the user's input text itself.

[0304] Step 3: Analyze emotional state

[0305] The device passes the received text to the sentiment analyzer. The sentiment analyzer uses, for example, the Hugging Face transformers library and analyzes the text with the BERT model. The input is the user's text received in step 2. The BERT model tokenizes the text and infers an emotion label based on the context. The analysis results indicate that the user is feeling anxious. The analysis results are sent to the server in JSON format.

[0306] Step 4: Parsing the request

[0307] The server analyzes the sentiment analysis results received in JSON format and the request text using natural language processing technology such as SpaCy. The input is the user's text and the sentiment analysis results. The text is tokenized using the SpaCy library, and key keywords such as "job change," "challenge," and "current workplace" are extracted. The analysis results become the input data for the next step.

[0308] Step 5: Selecting a generative AI model

[0309] The server selects the most suitable generative AI model based on the request content and the results of the sentiment analysis. The inputs are the keywords and emotional state extracted in step 4. A keyword matching algorithm is used to calculate the degree of match with the tags of each celebrity model and select the most suitable celebrity model. For example, based on the keywords "challenge" and "job change," a model that provides adventurous advice is selected. The results of this selection are the input data for the next step.

[0310] Step 6: Generating Advice

[0311] The server uses the selected generative AI model to generate advice appropriate to the user's request and emotional state. The input is the selected generative AI model and a prompt. Specifically, the prompt "I'm thinking about changing jobs. I'm feeling anxious. What should I think about this?" is input into GPT-3, and the server generates the advice "Identify where your passion lies and have the courage to take on challenges without fear of risk." This is the output data.

[0312] Step 7: Verify the advice

[0313] The server uses an internal algorithm to verify the appropriateness of the generated advice. The input is the advice generated in step 6. A rule-based filter is used to check whether it contains extreme expressions or inappropriate sentences. For example, regular expressions are used to check whether specific keywords are included and whether the expression is natural in the context. Adjustments are made as necessary, and finally, advice that passes the verification is generated as output data.

[0314] Step 8: Providing advice

[0315] The server finally sends the advice that has passed verification to the terminal and displays it to the user. The input is verified advice. The terminal uses HTML and CSS to display the received advice in an easy-to-read GUI (Graphical User Interface). Important points are highlighted to make it easier for the user to understand the advice.

[0316] (Application example 2)

[0317] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0318] Currently, there are limited ways to quickly provide appropriate advice to employees in brick-and-mortar stores regarding work-related or career-related issues. It is also difficult to provide advice that takes into account the emotional state of employees, resulting in problems such as a decline in employee motivation and the accumulation of stress. Furthermore, there is no established method for ensuring the reliability of the advice provided.

[0319] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving requests from users, means for saving famous people's quotes, personal experiences, and past advice as a dataset, means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset, means for generating advice in response to the user request using the selected generative AI model, an emotion engine for analyzing the user's emotional state, and an application for solving problems between employees and providing feedback. This makes it possible to quickly provide reliable advice that takes into account the emotional state of employees.

[0320] A "terminal" is a device that receives requests from users and provides generated advice.

[0321] "Server" means a central processing unit for storing and analyzing data, selecting and training generative AI models, and generating and providing advice.

[0322] A "celebrity" refers to a person who is widely known in society, and whose famous quotes, experiences, advice, etc. are used as a dataset.

[0323] A "dataset" is a collection of famous quotes, personal experiences, and past advice that is used as training material for generative AI models.

[0324] A "generative AI model" is an artificial intelligence model that generates advice in response to a user request based on a training dataset.

[0325] The "emotion engine" is software that analyzes the user's emotional state when they input a request and adjusts the advice content based on the results.

[0326] The "application for problem-solving and feedback among employees" is software that provides advice to employees in physical stores, taking into account their emotional state, in response to their concerns and questions.

[0327] "Internal Algorithm" means an internal computational means for validating the appropriateness of the generated advice and adjusting it if necessary.

[0328] A system for implementing the present invention includes the following elements:

[0329] Hardware and Software

[0330] Hardware: Smartphone

[0331] software:

[0332] Database: MySQL, PostgreSQL, etc.

[0333] Emotion engine: EmotionAPI

[0334] Generative AI model: OpenAI GPT-4 (registered trademark)

[0335] Application backend: Node.js

[0336] Frontend: React Native

[0337] Explanation of program processing

[0338] First, the server collects quotes, personal experiences, and past advice from famous people online and from books. This data is stored in a text database (MySQL or PostgreSQL) and cleaned under the supervision of experts. The cleaned data is organized into a training dataset, and a generative AI model (OpenAI GPT-4) is trained using this data.

[0339] Users access the application with specific concerns or questions and enter their request in text form via their smartphone. For example, they can enter something like, "I'm not sure whether I should apply for a new position or stay in my current one."

[0340] The device is equipped with an emotion engine (Emotion API) that analyzes the user's emotional state when they input a request. The emotion engine detects the user's emotions from the text input and sends the emotion analysis results to the server.

[0341] The server analyzes the received request and the results of sentiment analysis, and selects the most suitable generative AI model. This analysis is performed using syntactic analysis technology to extract keywords and important points contained in the request.

[0342] The server then uses the selected generative AI model to generate advice. The generative AI model references the learned thinking patterns and past advice of famous people to provide advice tailored to the user's concerns and emotional state. OpenAI GPT-4 is used to generate this advice.

[0343] The generated advice is then validated by an internal algorithm to ensure its appropriateness, including whether the advice is overly biased or if the wording is appropriate. If necessary, the advice is adjusted.

[0344] The final verified and adjusted advice is sent back to the terminal and provided to the user, who can then consider their own choices based on this advice.

[0345] Specific examples

[0346] Example 1: Career Choice and Sentiment Analysis

[0347] 1. The user enters the following text: "I'm not sure whether I should apply for a new position or stay in my current one."

[0348] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[0349] 3. The device sends the request and emotional state to the server.

[0350] 4. The server selects an appropriate generative AI model based on the request and the results of sentiment analysis.

[0351] 5. The server generates advice based on the generated AI model, such as "Identify your passion and have the courage to take on challenges without fear of risk," and adds expressions to ease anxiety.

[0352] 6. The server sends the verified advice to the terminal and displays it to the user.

[0353] Prompt Sentence Examples

[0354] "I'm not sure whether I should apply for a new position or stay in my current one. How should I decide? Helplessness: I feel it. Anxiety: I feel it. Confidence: I want it."

[0355] In this way, the system provides reliable advice that takes into account the user's emotional state, enabling store staff to solve problems and provide feedback.

[0356] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0357] Step 1:

[0358] Famous quotes, personal experiences, and past advice from famous people are collected online and from books. This data is stored in text format in a database (MySQL or PostgreSQL). The collected data is then cleaned and reviewed by experts and organized into a training dataset. This ensures high-quality data is used to train generative AI models.

[0359] Input: Text data collected from online sources and books

[0360] Output: Data organized as a training dataset

[0361] Step 2:

[0362] The prepared training dataset will be used to train a generative AI model (OpenAI GPT-4), which will then learn the thought patterns and advice of celebrities, laying the foundation for generating appropriate advice in response to user requests.

[0363] Input: A curated training dataset

[0364] Output: A trained generative AI model

[0365] Step 3:

[0366] The user enters specific concerns or questions in text format via their smartphone. For example, they might enter something like, "I'm not sure whether I should apply for a new position or stay in my current one." This text data is then analyzed in the next step.

[0367] Input: The textual request entered by the user

[0368] Output: Request data in text format

[0369] Step 4:

[0370] The emotion engine (Emotion API) built into the device analyzes the user's emotional state when they input a request. The emotion engine analyzes the text data and detects the emotion the user is feeling (e.g., anxiety or helplessness). The detected emotional state is sent to the server.

[0371] Input: Request data in text format

[0372] Output: Emotional analysis results (e.g. anxiety, helplessness)

[0373] Step 5:

[0374] The server analyzes the received request and the sentiment analysis results, and performs a syntactic analysis to extract keywords and key points contained in the request. This analysis clarifies the important points of the request content.

[0375] Input: Request data in text format, sentiment analysis results

[0376] Output: Keywords and key points of the request

[0377] Step 6:

[0378] The server selects the most suitable generative AI model based on the extracted keywords and key points, and uses the selected generative AI model to generate advice in response to the user's request. The generated advice also takes into account the user's emotional state.

[0379] Input: Keywords and key points of the request, sentiment analysis results

[0380] Output: Generated advice

[0381] Step 7:

[0382] The generated advice is verified by an algorithm inside the server to check its appropriateness, checking whether the advice is excessively biased and whether appropriate wording is used, and adjusting the advice content as necessary.

[0383] Input: Generated advice

[0384] Output: Validated and adjusted advice

[0385] Step 8:

[0386] The final verified and adjusted advice is then sent back to the device and provided to the user, who can then receive the advice via their smartphone and consider their own choices.

[0387] Input: Validated and adjusted advice

[0388] Output: Advice given to the user

[0389] The above steps make it possible to provide fast, reliable advice that takes into account the emotional state of employees in physical stores in response to their concerns and questions.

[0390] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0391] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0392] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0393] [Second embodiment]

[0394] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0395] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0396] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0397] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0398] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0399] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0400] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0401] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0402] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0403] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0404] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0405] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0406] The present invention is a system that enables users to obtain reliable advice to solve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores famous quotes, personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models.

[0407] Program processing

[0408] Data Collection and Training

[0409] The server collects quotes, personal experiences, and past advice from famous people online and in books. The collected data is saved in text format and then edited by experts to be ready for use as a training dataset. This dataset is used to train a generative AI model. The trained model reflects the thinking patterns and advice of famous people.

[0410] Receiving and parsing the request

[0411] The user accesses the system and inputs specific concerns or questions into the terminal. For example, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[0412] AI model selection and generation

[0413] The server analyzes the received request and selects the celebrity generative AI model that best suits the request. The selection criteria are the relevance between the request content and the training data. The selected generative AI model is used to generate advice that meets the user's request. The generation process ensures that reliable advice is provided based on the celebrity's thought patterns and past advice that the model has learned.

[0414] Providing advice

[0415] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is not overly biased and whether it is expressed in an appropriate context. If the verified advice is deemed appropriate, it is sent to the terminal and displayed to the user. The user can then consider their own choices and make a decision based on this advice.

[0416] Specific examples

[0417] Example 1: Career Choices

[0418] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0419] 2. The device receives this request and sends it to the server.

[0420] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model with extensive experience in career decisions.

[0421] 4. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks."

[0422] 5. The server sends the verified advice to the terminal and displays it to the user.

[0423] Example 2: Academic direction

[0424] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0425] 2. The device receives this request and sends it to the server.

[0426] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model that is good at giving academic advice.

[0427] 4. Based on the AI ​​model, the server generates advice such as, "Value your own interests and curiosity, and make a choice based on which one is best suited to you."

[0428] 5. The server sends the verified advice to the terminal and displays it to the user.

[0429] This system allows users to obtain reliable advice that can be used effectively in making life choices.

[0430] The processing flow will be explained below.

[0431] Step 1:

[0432] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[0433] Step 2:

[0434] The server uses the training dataset to train the generative AI model. Specifically, it uses natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[0435] Step 3:

[0436] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[0437] Step 4:

[0438] The terminal receives a request input by the user and transmits the content in text format to the server, which then formats the request appropriately for the system.

[0439] Step 5:

[0440] The server analyzes the received request. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable AI model to generate a celebrity.

[0441] Step 6:

[0442] The server uses the selected generative AI model to generate advice for the user's request, referencing the thinking patterns and past advice of famous people that the model has learned, and generates advice that is specifically suited to the request.

[0443] Step 7:

[0444] The server uses an internal algorithm to verify the appropriateness of the advice it generates, for example, by checking that the advice is not overly biased and that it uses appropriate language in the context. If necessary, it adjusts the advice it generates.

[0445] Step 8:

[0446] The server then sends the final verified and adjusted advice to the terminal, where it is displayed to the user.

[0447] Step 9:

[0448] Users receive advice displayed on their devices and use it to consider their choices, helping them make better decisions about their life paths.

[0449] In this way, the system provides users with reliable advice and allows them to obtain useful information in making life choices.

[0450] Example 1

[0451] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0452] In today's world, when individuals face various life choices, it is difficult to obtain advice based on reliable information. Furthermore, existing advice systems are general and cannot provide specific advice tailored to individual concerns or questions. Therefore, there is a need for the development of a system that can address the specific problems faced by individual users and provide accurate advice based on deep insight and experience.

[0453] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0454] In this invention, the server includes an information processing device that receives requests from users, a data storage device that uses a database to store quotes and personal experiences, an analysis device that analyzes the received requests using a natural language processing tool, a learning device that selects and trains an optimal generative AI model based on the analysis results, a generation device that generates advice in response to the user request using the selected generative AI model, and a display device that provides the generated advice to the user. This makes it possible to provide highly reliable and specific advice based on the concerns and questions of individual users.

[0455] An "information processing device" is a device capable of receiving input from a user and forwarding it to other system components.

[0456] A "data storage device" is a device that stores data such as famous quotes and personal experiences in a database and has the function of retrieving that data as needed.

[0457] An "analysis device" is a device that uses natural language processing tools to analyze a received request and has the function of understanding the content of the request.

[0458] A "learning device" is a device that has the function of selecting the optimal generative AI model based on the analysis results and training that model.

[0459] A "generation device" is a device that has the function of generating specific advice in response to a user request using a selected generative AI model.

[0460] A "display device" is a device that has the function of providing and displaying the generated advice to the user.

[0461] A "natural language processing tool" is software or a library that analyzes received text data and understands and classifies its content.

[0462] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task based on a training dataset.

[0463] The present invention is a system that allows users to get reliable advice to solve specific questions and concerns about their life path. The system includes the following main components:

[0464] Data collection

[0465] The server collects quotes, personal experiences, and past advice from famous people from the internet and books. For this collection, it uses web scraping tools such as Scrapy and Beautiful Soup. The collected data is then stored in a database in text format.

[0466] Data Preparation and Training

[0467] The server then uses Hugging Face's Transformers library to organize the collected text data into a training dataset, allowing the generative AI model to learn the thought patterns and advice of celebrities. This organization process also uses natural language processing libraries such as NLTK (Natural Language Toolkit) and spaCy.

[0468] Receiving a user request

[0469] Users access the system using a dedicated app or a web browser and enter their specific concerns or questions. The text is then sent to the server via web application frameworks such as Flask and Django.

[0470] Parsing the request

[0471] The server uses spaCy to perform natural language analysis on the received request. Based on the analysis results, the request content is classified and an appropriate generative AI model is selected. The selection criterion in this case is the relevance of the request content to the training dataset.

[0472] Generating Advice

[0473] The server uses the selected generative AI model to generate specific advice based on the user's request. The generation process incorporates the thought patterns and past advice of famous people that have been trained in advance.

[0474] Validating and providing advice

[0475] The generated advice is then validated using an internal algorithm to ensure it is appropriate. Specifically, it checks whether the advice is overly biased and whether it is expressed appropriately for the context. After this validation is complete, the server sends the advice to the terminal via Flask or Django and displays it to the user.

[0476] Specific examples

[0477] Example 1: Career Choices

[0478] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0479] 2. The device receives this request and sends it to the server via Flask.

[0480] 3. The server uses spaCy to analyze the request and extract keywords such as "job change" and "workplace."

[0481] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as "Pursue your passion and take risks."

[0482] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[0483] Example 2: Academic direction

[0484] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0485] 2. The device receives this request and sends it to the server via Flask.

[0486] 3. The server uses spaCy to analyze the request and extract keywords such as "science" and "humanities."

[0487] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as, "Value your interests and curiosity, and make a choice based on which one is most suited to you."

[0488] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[0489] As described above, this system allows users to obtain reliable and specific advice, and the entire system operates through a consistent process, providing a fast and appropriate response to user requests.

[0490] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0491] Step 1:

[0492] Users access the system using a dedicated app or a web browser and enter specific concerns or questions. The entered text is sent to the server via Flask or Django.

[0493] Specific behavior: The user enters the following into the web form: "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job," and clicks the submit button. This is the input.

[0494] Step 2:

[0495] The terminal receives this request and sends it to the server as text data. The server receives the request.

[0496] Specific operation: The terminal receives input and sends a POST request to the server via Flask or Django in JSON or text format, which is the output to the server.

[0497] Step 3:

[0498] The request received by the server is analyzed. spaCy, a natural language processing tool, is used to understand the content of the request and extract keywords. This is the input. Based on the analysis results, the content of the request is classified.

[0499] Specific operation: The server analyzes a request such as "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job" using spaCy, and extracts keywords such as "job change" and "workplace." This is the output of the analysis.

[0500] Step 4:

[0501] The server selects and trains the optimal generative AI model based on the analysis results. The selection criteria are the relevance of the request content to the training dataset, which is the input. A learning device is used to select the model and conduct additional training if necessary.

[0502] Specific operation: The server selects a generative AI model for "Career Advisor" based on the analyzed keywords, and trains the model for the number of epochs using the training dataset. This is the output of model selection and training.

[0503] Step 5:

[0504] The server uses the selected generative AI model to generate specific advice in response to the user's request. The input is the user's request and the selected AI model. The generation process reflects the thinking patterns and past advice of famous people that have been trained in advance.

[0505] How it works: The server inputs a prompt such as "I'm thinking about changing jobs, but should I try a new job?" into the career advisor's generative AI model, and gets the generated advice "Follow your passion and take risks." This is the output of the generation.

[0506] Step 6:

[0507] The server validates the appropriateness of the generated advice using an internal algorithm. The input is the generated advice. It checks whether the advice is overly biased and whether it is expressed appropriately for the context.

[0508] What happens: The server validates the generated advice, "Follow your passion and take risks," to check for inappropriate content. This is the validation output:

[0509] Step 7:

[0510] If the server determines that the advice is valid, it sends it to the terminal via Flask or Django and displays it to the user. The input is the validated advice.

[0511] What happens: The server sends the advice "Follow your passion and take risks" to the terminal through the Flask and Django application and displays it on the user's screen. This is the display output:

[0512] (Application example 1)

[0513] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0514] In traditional brick-and-mortar stores, it was difficult for customers to ask questions about specific products or services in real time and receive reliable advice. There were also few ways to effectively utilize advice based on the words and personal experiences of celebrities. As a result, customers lacked the information they needed to make the best choice for themselves, leading to problems such as reduced purchasing efficiency and satisfaction.

[0515] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0516] In this invention, the server includes a means for receiving requests from users, a means for storing famous people's quotes, personal experiences, and past advice as a dataset, and a means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset. This allows customers to receive product advice in real time in a physical store. Furthermore, the selected generative AI model can be used to provide reliable advice, quickly and accurately providing customers with information to help them make the best choice for themselves.

[0517] A "terminal that receives requests from a user" is a device through which a user inputs questions or requests and transmits that information to other system components.

[0518] An "information processing device that stores the words, experiences, and past advice of famous people as a data set" is a device that collects, organizes, and stores data such as the statements and experiences of famous people.

[0519] An "information processing device that analyzes received requests and selects and trains a generative AI model based on the required training data set" is a device that analyzes requests from users, selects an AI model based on appropriate training data, and trains it.

[0520] An "information processing device that generates advice in response to a user request using a selected generative AI model" is a device that generates specific advice in response to a user question using a selected AI model.

[0521] "Means that enable customers to obtain product advice using a terminal in a physical store" refers to a system that enables customers to obtain advice on product selection and purchases through a terminal in a physical store environment.

[0522] The "terminal for providing generated advice to a user" is a device for displaying, notifying, or communicating generated advice to a user.

[0523] "Formalizing the dataset with the permission and supervision of the celebrity" means recording the statements and activities of the celebrity as official data with the celebrity's permission and supervision.

[0524] A "means for generating reliable advice" is a process or mechanism for providing accurate and useful advice to users based on reliable sources of information and data.

[0525] "Verifying the appropriateness of the generated advice using an internal algorithm" means confirming the accuracy and appropriateness of the advice generated using an internal algorithm.

[0526] "Means for adjustment as necessary" refers to a mechanism for correcting or improving the content of advice generated if it is inappropriate.

[0527] This invention is a system that allows customers to ask questions about product selection and purchases in a physical store and receive reliable advice in real time. This system consists of a terminal where the user inputs their questions, an information processing device (server) that stores and utilizes quotes and personal experiences of celebrities and past advice as a dataset, and an information processing device that trains and executes a generative AI model.

[0528] Hardware and Software

[0529] Hardware:

[0530] Smartphones, smart glasses (to input user questions and display advice)

[0531] Servers (for storing data, training and running AI models)

[0532] software:

[0533] OpenAI API (AI model generation and execution)

[0534] HTTP request library (requests)

[0535] Program processing overview

[0536] The server collects the words, experiences, and past advice of celebrities from online and offline sources and stores them as a dataset. The collected data is then used to train a generative AI model. This model learns the thought patterns and advice of celebrities and generates advice accordingly.

[0537] When a user enters a question using a smartphone or smart glasses in a physical store, the device sends the request in text format to the server. The server analyzes the request, selects the most appropriate generative AI model, and generates advice. The generated advice is verified for appropriateness by an internal algorithm, and if deemed appropriate, it is sent to the device and displayed to the user.

[0538] Specific examples

[0539] Selecting products in the store

[0540] A customer in a physical store puts on smart glasses and inputs a question such as, "Which brand is best when choosing new sneakers?"

[0541] Example prompt sentence:

[0542] Customer Question: What brand is best when choosing new sneakers?

[0543] You are an AI that provides advice based on the experiences and words of famous athletes. Please advise users on which brand of sneakers to choose by referring to the following dataset:

[0544] Dataset:

[0545] 1. Comments from famous athletes about their sneaker choices

[0546] 2. Expert-recommended brands

[0547] 3. Industry magazine rankings

[0548] This system allows customers to instantly receive helpful advice on product selection in-store and receive assistance in making the best choices.

[0549] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0550] Step 1:

[0551] The server collects quotes, personal experiences, and past advice from famous people both online and offline. The collected data is stored in text format and then compiled into training data through expert supervision. This allows the server to maintain a highly reliable data set.

[0552] Input: Text data collected from the internet and books

[0553] Output: A curated training dataset curated by experts

[0554] Specific operations: Data is obtained from online databases and books using web scraping and OCR technology, and then reviewed and tagged by experts.

[0555] Step 2:

[0556] The server trains the generative AI model using a curated training dataset, which learns the thought patterns and advice of celebrities.

[0557] Input: Training dataset

[0558] Output: A trained generative AI model

[0559] What it does: Uses a training dataset to train a model in an AI model training framework (e.g., TensorFlow, PyTorch).

[0560] Step 3:

[0561] Users use their smartphones or smart glasses in a physical store to input questions into the terminal, which then sends the questions in text format to the server.

[0562] Input: User question text

[0563] Output: Question text sent to the server

[0564] Specific operation: The user enters a question into the application's interface, which is then sent by the device to the server.

[0565] Step 4:

[0566] The server analyzes the received request and selects the most suitable generative AI model based on the relevance of the request content to the training data.

[0567] Input: User question text

[0568] Output: The selected generative AI model

[0569] Specific operation: The server uses natural language processing technology to analyze the question text and select the corresponding generative AI model.

[0570] Step 5:

[0571] The server uses the selected generative AI model to generate advice for the user's question. The generated advice is based on the thought patterns and past advice of famous people that the model has learned.

[0572] Input: User question text, generative AI model

[0573] Output: The generated advice

[0574] Specific operation: The server inputs the prompt sentence into the selected generative AI model and performs text generation using AI.

[0575] Step 6:

[0576] The server uses an internal algorithm to verify the appropriateness of the advice it generates, ensuring that it is expressed in the right context and is not overly biased.

[0577] Input: Generated advice

[0578] Output: Verified advice

[0579] What happens: The server uses internal algorithms and rule-based validation systems to check the content of the advice generated.

[0580] Step 7:

[0581] The server sends the verified advice to the user's device, which displays it to the user, who then considers his or her options and makes a decision based on the advice displayed.

[0582] Input: Verified Advice

[0583] Output: Advice displayed by the user

[0584] Specific operation: The server sends the advice in an appropriate format to the terminal, and the terminal displays the advice on the screen.

[0585] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0586] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components include a device that receives user requests, a server that stores famous quotes and personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models. The system also incorporates an emotion engine that recognizes the user's emotions and can analyze the user's emotional state when the request is made and tailor the advice accordingly.

[0587] Program processing

[0588] Data Collection and Training

[0589] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and edited by experts to create a training dataset. A generative AI model is trained using this dataset. The trained model reflects the thinking patterns and advice of famous people.

[0590] Receiving requests and analyzing sentiment

[0591] The user accesses the system and inputs specific concerns or questions in text format from a terminal. For example, they might input something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[0592] Analysis by emotion engine

[0593] The device incorporates an emotion engine that recognizes the user's emotional state when inputting a request. The emotion engine analyzes the user's emotions through analysis of the user's voice input and text, and sends the analysis results to the server. For example, if the user is feeling anxious, that emotional state will be included.

[0594] Request and emotional state analysis

[0595] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model and generates advice taking into account the user's emotional state.

[0596] AI model selection and generation

[0597] The server uses the selected generative AI model to generate advice based on the user's request and emotional state, and provides advice that matches the user's emotional state by referencing the thought patterns and past advice of famous people that the model has learned.

[0598] Validating and providing advice

[0599] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is overly biased and whether it uses appropriate language in line with the context. If necessary, it adjusts the generated advice. If the final verified and adjusted advice is deemed appropriate, it is sent to the device and displayed to the user. The user can then consider their own choices based on this advice.

[0600] Specific examples

[0601] Example 1: Career Choice and Sentiment Analysis

[0602] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0603] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[0604] 3. The device sends the request and emotional state to the server.

[0605] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[0606] 5. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks," and adds phrases to ease anxiety.

[0607] 6. The server sends the verified advice to the terminal and displays it to the user.

[0608] Example 2: Academic orientation and sentiment analysis

[0609] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0610] 2. The device receives this request and uses its emotion engine to analyze that the user is excited.

[0611] 3. The device sends the request and emotional state to the server.

[0612] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[0613] 5. Based on the AI ​​model, the server generates advice such as "Value your interests and curiosity, and make a choice based on which direction they are directed," and adds expressions to help the user stay calm.

[0614] 6. The server sends the verified advice to the terminal and displays it to the user.

[0615] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[0616] The processing flow will be explained below.

[0617] Step 1:

[0618] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[0619] Step 2:

[0620] The server uses the training dataset to train the generative AI model, specifically using natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[0621] Step 3:

[0622] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[0623] Step 4:

[0624] The device receives requests entered by the user and sends them in text format to the server. The device's built-in emotion engine analyzes the user's emotional state as they speak, extracting emotional states such as joy, sadness, anxiety, and excitement based on the user's voice input and text information.

[0625] Step 5:

[0626] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model. The selection criteria are the relevance of the request content to the training data and the user's emotional state.

[0627] Step 6:

[0628] The server uses the selected generative AI model to generate advice in response to the user's request. The generated advice is based on the thought patterns and past advice of famous people that the model has learned. In addition, the expression and content are adjusted according to the user's emotional state. For example, if a user is feeling anxious, reassuring expressions are used.

[0629] Step 7:

[0630] The server uses an internal algorithm to verify the appropriateness of the generated advice. It checks whether the advice is excessively biased and whether appropriate expressions are used for the context. If necessary, it makes corrections to the generated advice.

[0631] Step 8:

[0632] The server then sends the final verified and corrected advice to the terminal, which then displays it to the user.

[0633] Step 9:

[0634] Users receive advice displayed on their devices and use it to consider their own choices. For example, a user receiving advice about changing jobs can use the advice to determine the direction they truly want to take. Furthermore, advice appropriately adjusted by the emotion engine is tailored to the user's emotional state, allowing them to accept the advice with greater confidence.

[0635] In this way, the system provides users with reliable advice that takes into account their emotional state, enabling them to obtain useful information for making life choices.

[0636] Example 2

[0637] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0638] In modern society, many people are concerned about their careers, studies, and life paths. However, it is not easy to get reliable advice. In addition, there are limited advice-giving systems that take the user's emotional state into account, and general advice often lacks emotional consideration. This makes it difficult for users to make appropriate and meaningful decisions.

[0639] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a terminal that receives a request from a user, a means for saving famous quotes, personal experiences, and past advice as a dataset, a means for analyzing the received request and extracting necessary keywords and key points using natural language processing technology, a sentiment analysis means that recognizes the user's emotional state and adjusts the content of the advice based on the analysis results, a means for selecting and training a generative AI model based on a necessary training dataset, a means for generating advice in response to the user's request using the selected generative AI model, a means for verifying the appropriateness of the generated advice using an internal algorithm and adjusting it as necessary, and a terminal that provides the generated advice to the user. This makes it possible to provide highly reliable advice that takes the user's emotional state into consideration.

[0640] "User" refers to a person who accesses the system, enters their concerns and questions, and seeks advice.

[0641] "Terminal" means a device that allows a user to access the system and input / submit requests. Examples include PCs and smartphones.

[0642] A "quote" is a commonly quoted phrase or expression uttered by a famous or influential individual.

[0643] "Testimonials" refer to stories or accounts recorded by individuals about specific events or experiences.

[0644] "Past advice" refers to advice or suggestions given to others in the past.

[0645] "Dataset" refers to a set of data collected and organized for use in the System.

[0646] A "server" is a computer system that receives requests from users, processes them, and generates and provides advice.

[0647] "Natural language processing technology" refers to the techniques and algorithms that enable computers to understand and process human language.

[0648] "Keywords" refer to words or phrases that play a significant role in a user's request or text.

[0649] "Emotional state" refers to the emotions or psychological state a user experiences in a particular situation.

[0650] "Emotion analysis means" refers to the technology or algorithm that recognizes emotions from user requests or voice and provides the analysis results.

[0651] A "generative AI model" is an artificial intelligence model that is trained using a large dataset and is used to generate advice in response to a user request.

[0652] "Relevance" refers to the unbiased nature of the advice generated, and whether it is natural, contextual, and appropriate.

[0653] "Internal algorithm" refers to the calculation methods and programs used within the system.

[0654] "Validation" refers to the process of evaluating whether the generated advice is appropriate.

[0655] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores quotes, personal experiences, and past advice as a dataset, and a server that trains and executes a generative AI model. The system also incorporates an emotion analysis means that recognizes the user's emotions, and can analyze the user's emotional state when the request is made and adjust the content of the advice accordingly.

[0656] The server uses Python scripts and web analytics tools to collect quotes and testimonials from online sources and books, and store them in a database in text format. Specifically, the Beautiful Soup and Requests libraries are used to scrape data from websites, the Pandas library is used to format and clean the data, and then the data is stored in a MySQL database via SQLAlchemy.

[0657] The collected dataset is used to train a generative AI model (e.g., GPT-3) using GPUs on cloud computing platforms (e.g., AWS or Google Cloud). The training process is carried out over hundreds of epochs using libraries such as TensorFlow and PyTorch.

[0658] Users access the web application using a device such as a PC or smartphone and enter specific concerns or questions in text format. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The device then asynchronously sends this request to the server using JavaScript (AJAX).

[0659] The device uses emotion analysis to recognize the user's emotional state from the input text. For example, it uses the BERT model with the Hugging Face transformers library to classify emotions and obtain emotion labels such as "anxiety" or "excitement." The analysis results are sent to the server in JSON format.

[0660] The server combines the request and the emotional state and uses natural language processing (NLP) techniques to analyze the request. For example, it uses the spaCy library to tokenize the text, extract nouns and verbs, and identify keywords. Keywords such as "job change," "new workplace," and "challenge" are extracted.

[0661] The server selects the most suitable generative AI model based on the analysis results. It uses a keyword matching algorithm to select the best one from multiple celebrity models. If the keyword "challenge" is included, it selects a model that provides adventurous advice.

[0662] Using the selected generative AI model, the server generates advice based on the user's request and emotional state. For example, using GPT-3, the server generates advice such as, "Identify your passion and have the courage to take risks." The prompt used is, "I'm thinking about changing jobs. I'm feeling anxious. What should I do?"

[0663] The server uses internal algorithms to verify the appropriateness of the generated advice and adjusts it as needed, for example by using rule-based filters to ensure that it does not contain extreme or inappropriate language. Finally, advice that passes the verification is sent to the device and displayed to the user.

[0664] Specific examples of prompts are as follows:

[0665] "I'm thinking about changing jobs, but I'm not sure whether I should stay where I am or try something new. What should I do? I'm feeling anxious."

[0666] "I'm worried about my further education. I'm not sure whether to go into science or humanities, but I'm feeling excited. I'd appreciate some advice."

[0667] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[0668] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0669] Step 1: User request input

[0670] A user enters specific questions or concerns about their life path into a text box in a web application. For example, they might enter, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job." This is the input data. The device asynchronously sends the text entered by the user to the server using AJAX. This sent data is the input for the next step.

[0671] Step 2: Receiving the request

[0672] The terminal receives text input from the user and passes it to the emotion analysis means. The input data is the text information sent by the user. The server receives the text received via an AJAX request for analysis. The output is the user's input text itself.

[0673] Step 3: Analyze emotional state

[0674] The device passes the received text to the sentiment analyzer. The sentiment analyzer uses, for example, the Hugging Face transformers library and analyzes the text with the BERT model. The input is the user's text received in step 2. The BERT model tokenizes the text and infers an emotion label based on the context. The analysis results indicate that the user is feeling anxious. The analysis results are sent to the server in JSON format.

[0675] Step 4: Parsing the request

[0676] The server analyzes the sentiment analysis results received in JSON format and the request text using natural language processing technology such as SpaCy. The input is the user's text and the sentiment analysis results. The text is tokenized using the SpaCy library, and key keywords such as "job change," "challenge," and "current workplace" are extracted. The analysis results become the input data for the next step.

[0677] Step 5: Selecting a generative AI model

[0678] The server selects the most suitable generative AI model based on the request content and the results of the sentiment analysis. The inputs are the keywords and emotional state extracted in step 4. A keyword matching algorithm is used to calculate the degree of match with the tags of each celebrity model and select the most suitable celebrity model. For example, based on the keywords "challenge" and "job change," a model that provides adventurous advice is selected. The results of this selection are the input data for the next step.

[0679] Step 6: Generating Advice

[0680] The server uses the selected generative AI model to generate advice appropriate to the user's request and emotional state. The input is the selected generative AI model and a prompt. Specifically, the prompt "I'm thinking about changing jobs. I'm feeling anxious. What should I think about this?" is input into GPT-3, and the server generates the advice "Identify where your passion lies and have the courage to take on challenges without fear of risk." This is the output data.

[0681] Step 7: Verify the advice

[0682] The server uses an internal algorithm to verify the appropriateness of the generated advice. The input is the advice generated in step 6. A rule-based filter is used to check whether it contains extreme expressions or inappropriate sentences. For example, regular expressions are used to check whether specific keywords are included and whether the expression is natural in the context. Adjustments are made as necessary, and finally, advice that passes the verification is generated as output data.

[0683] Step 8: Providing advice

[0684] The server finally sends the advice that has passed verification to the terminal and displays it to the user. The input is verified advice. The terminal uses HTML and CSS to display the received advice in an easy-to-read GUI (Graphical User Interface). Important points are highlighted to make it easier for the user to understand the advice.

[0685] (Application example 2)

[0686] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0687] Currently, there are limited ways to quickly provide appropriate advice to employees in brick-and-mortar stores regarding work-related or career-related issues. It is also difficult to provide advice that takes into account the emotional state of employees, resulting in problems such as a decline in employee motivation and the accumulation of stress. Furthermore, there is no established method for ensuring the reliability of the advice provided.

[0688] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving requests from users, means for saving famous people's quotes, personal experiences, and past advice as a dataset, means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset, means for generating advice in response to the user request using the selected generative AI model, an emotion engine for analyzing the user's emotional state, and an application for solving problems between employees and providing feedback. This makes it possible to quickly provide reliable advice that takes into account the emotional state of employees.

[0689] A "terminal" is a device that receives requests from users and provides generated advice.

[0690] "Server" means a central processing unit for storing and analyzing data, selecting and training generative AI models, and generating and providing advice.

[0691] A "celebrity" refers to a person who is widely known in society, and whose famous quotes, experiences, advice, etc. are used as a dataset.

[0692] A "dataset" is a collection of famous quotes, personal experiences, and past advice that is used as training material for generative AI models.

[0693] A "generative AI model" is an artificial intelligence model that generates advice in response to a user request based on a training dataset.

[0694] The "emotion engine" is software that analyzes the user's emotional state when they input a request and adjusts the advice content based on the results.

[0695] The "application for problem-solving and feedback among employees" is software that provides advice to employees in physical stores, taking into account their emotional state, in response to their concerns and questions.

[0696] "Internal Algorithm" means an internal computational means for validating the appropriateness of the generated advice and adjusting it if necessary.

[0697] A system for implementing the present invention includes the following elements:

[0698] Hardware and Software

[0699] Hardware: Smartphone

[0700] software:

[0701] Database: MySQL, PostgreSQL, etc.

[0702] Emotion engine: EmotionAPI

[0703] Generative AI model: OpenAI GPT-4

[0704] Application backend: Node.js

[0705] Frontend: React Native

[0706] Explanation of program processing

[0707] First, the server collects quotes, personal experiences, and past advice from famous people online and from books. This data is stored in a text database (MySQL or PostgreSQL) and cleaned under the supervision of experts. The cleaned data is organized into a training dataset, and a generative AI model (OpenAI GPT-4) is trained using this data.

[0708] Users access the application with specific concerns or questions and enter their request in text form via their smartphone. For example, they can enter something like, "I'm not sure whether I should apply for a new position or stay in my current one."

[0709] The device is equipped with an emotion engine (Emotion API) that analyzes the user's emotional state when they input a request. The emotion engine detects the user's emotions from the text input and sends the emotion analysis results to the server.

[0710] The server analyzes the received request and the results of sentiment analysis, and selects the most suitable generative AI model. This analysis is performed using syntactic analysis technology to extract keywords and important points contained in the request.

[0711] The server then uses the selected generative AI model to generate advice. The generative AI model references the learned thinking patterns and past advice of famous people to provide advice tailored to the user's concerns and emotional state. OpenAI GPT-4 is used to generate this advice.

[0712] The generated advice is then validated by an internal algorithm to ensure its appropriateness, including whether the advice is overly biased or if the wording is appropriate. If necessary, the advice is adjusted.

[0713] The final verified and adjusted advice is sent back to the terminal and provided to the user, who can then consider their own choices based on this advice.

[0714] Specific examples

[0715] Example 1: Career Choice and Sentiment Analysis

[0716] 1. The user enters the following text: "I'm not sure whether I should apply for a new position or stay in my current one."

[0717] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[0718] 3. The device sends the request and emotional state to the server.

[0719] 4. The server selects an appropriate generative AI model based on the request and the results of sentiment analysis.

[0720] 5. The server generates advice based on the generated AI model, such as "Identify your passion and have the courage to take on challenges without fear of risk," and adds expressions to ease anxiety.

[0721] 6. The server sends the verified advice to the terminal and displays it to the user.

[0722] Prompt Sentence Examples

[0723] "I'm not sure whether I should apply for a new position or stay in my current one. How should I decide? Helplessness: I feel it. Anxiety: I feel it. Confidence: I want it."

[0724] In this way, the system provides reliable advice that takes into account the user's emotional state, enabling store staff to solve problems and provide feedback.

[0725] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0726] Step 1:

[0727] Famous quotes, personal experiences, and past advice from famous people are collected online and from books. This data is stored in text format in a database (MySQL or PostgreSQL). The collected data is then cleaned and reviewed by experts and organized into a training dataset. This ensures high-quality data is used to train generative AI models.

[0728] Input: Text data collected from online sources and books

[0729] Output: Data organized as a training dataset

[0730] Step 2:

[0731] The prepared training dataset will be used to train a generative AI model (OpenAI GPT-4), which will then learn the thought patterns and advice of celebrities, laying the foundation for generating appropriate advice in response to user requests.

[0732] Input: A curated training dataset

[0733] Output: A trained generative AI model

[0734] Step 3:

[0735] The user enters specific concerns or questions in text format via their smartphone. For example, they might enter something like, "I'm not sure whether I should apply for a new position or stay in my current one." This text data is then analyzed in the next step.

[0736] Input: The textual request entered by the user

[0737] Output: Request data in text format

[0738] Step 4:

[0739] The emotion engine (Emotion API) built into the device analyzes the user's emotional state when they input a request. The emotion engine analyzes the text data and detects the emotion the user is feeling (e.g., anxiety or helplessness). The detected emotional state is sent to the server.

[0740] Input: Request data in text format

[0741] Output: Emotional analysis results (e.g. anxiety, helplessness)

[0742] Step 5:

[0743] The server analyzes the received request and the sentiment analysis results, and performs a syntactic analysis to extract keywords and key points contained in the request. This analysis clarifies the important points of the request content.

[0744] Input: Request data in text format, sentiment analysis results

[0745] Output: Keywords and key points of the request

[0746] Step 6:

[0747] The server selects the most suitable generative AI model based on the extracted keywords and key points, and uses the selected generative AI model to generate advice in response to the user's request. The generated advice also takes into account the user's emotional state.

[0748] Input: Keywords and key points of the request, sentiment analysis results

[0749] Output: Generated advice

[0750] Step 7:

[0751] The generated advice is verified by an algorithm inside the server to check its appropriateness, checking whether the advice is excessively biased and whether appropriate wording is used, and adjusting the advice content as necessary.

[0752] Input: Generated advice

[0753] Output: Validated and adjusted advice

[0754] Step 8:

[0755] The final verified and adjusted advice is then sent back to the device and provided to the user, who can then receive the advice via their smartphone and consider their own choices.

[0756] Input: Validated and adjusted advice

[0757] Output: Advice given to the user

[0758] The above steps make it possible to provide fast, reliable advice that takes into account the emotional state of employees in physical stores in response to their concerns and questions.

[0759] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0760] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0761] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0762] [Third embodiment]

[0763] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0764] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0765] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0766] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0767] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0768] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0769] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0770] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0771] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0772] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0773] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0774] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. 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."

[0775] The present invention is a system that enables users to obtain reliable advice to solve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores famous quotes, personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models.

[0776] Program processing

[0777] Data Collection and Training

[0778] The server collects quotes, personal experiences, and past advice from famous people online and in books. The collected data is saved in text format and then edited by experts to be ready for use as a training dataset. This dataset is used to train a generative AI model. The trained model reflects the thinking patterns and advice of famous people.

[0779] Receiving and parsing the request

[0780] The user accesses the system and inputs specific concerns or questions into the terminal. For example, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[0781] AI model selection and generation

[0782] The server analyzes the received request and selects the celebrity generative AI model that best suits the request. The selection criteria are the relevance between the request content and the training data. The selected generative AI model is used to generate advice that meets the user's request. The generation process ensures that reliable advice is provided based on the celebrity's thought patterns and past advice that the model has learned.

[0783] Providing advice

[0784] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is not overly biased and whether it is expressed in an appropriate context. If the verified advice is deemed appropriate, it is sent to the terminal and displayed to the user. The user can then consider their own choices and make a decision based on this advice.

[0785] Specific examples

[0786] Example 1: Career Choices

[0787] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0788] 2. The device receives this request and sends it to the server.

[0789] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model with extensive experience in career decisions.

[0790] 4. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks."

[0791] 5. The server sends the verified advice to the terminal and displays it to the user.

[0792] Example 2: Academic direction

[0793] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0794] 2. The device receives this request and sends it to the server.

[0795] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model that is good at giving academic advice.

[0796] 4. Based on the AI ​​model, the server generates advice such as, "Value your own interests and curiosity, and make a choice based on which one is best suited to you."

[0797] 5. The server sends the verified advice to the terminal and displays it to the user.

[0798] This system allows users to obtain reliable advice that can be used effectively in making life choices.

[0799] The processing flow will be explained below.

[0800] Step 1:

[0801] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[0802] Step 2:

[0803] The server uses the training dataset to train the generative AI model. Specifically, it uses natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[0804] Step 3:

[0805] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[0806] Step 4:

[0807] The terminal receives a request input by the user and transmits the content in text format to the server, which then formats the request appropriately for the system.

[0808] Step 5:

[0809] The server analyzes the received request. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable AI model to generate a celebrity.

[0810] Step 6:

[0811] The server uses the selected generative AI model to generate advice for the user's request, referencing the thinking patterns and past advice of famous people that the model has learned, and generates advice that is specifically suited to the request.

[0812] Step 7:

[0813] The server uses an internal algorithm to verify the appropriateness of the advice it generates, for example, by checking that the advice is not overly biased and that it uses appropriate language in the context. If necessary, it adjusts the advice it generates.

[0814] Step 8:

[0815] The server then sends the final verified and adjusted advice to the terminal, where it is displayed to the user.

[0816] Step 9:

[0817] Users receive advice displayed on their devices and use it to consider their choices, helping them make better decisions about their life paths.

[0818] In this way, the system provides users with reliable advice and allows them to obtain useful information in making life choices.

[0819] Example 1

[0820] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0821] In today's world, when individuals face various life choices, it is difficult to obtain advice based on reliable information. Furthermore, existing advice systems are general and cannot provide specific advice tailored to individual concerns or questions. Therefore, there is a need for the development of a system that can address the specific problems faced by individual users and provide accurate advice based on deep insight and experience.

[0822] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0823] In this invention, the server includes an information processing device that receives requests from users, a data storage device that uses a database to store quotes and personal experiences, an analysis device that analyzes the received requests using a natural language processing tool, a learning device that selects and trains an optimal generative AI model based on the analysis results, a generation device that generates advice in response to the user request using the selected generative AI model, and a display device that provides the generated advice to the user. This makes it possible to provide highly reliable and specific advice based on the concerns and questions of individual users.

[0824] An "information processing device" is a device capable of receiving input from a user and forwarding it to other system components.

[0825] A "data storage device" is a device that stores data such as famous quotes and personal experiences in a database and has the function of retrieving that data as needed.

[0826] An "analysis device" is a device that uses natural language processing tools to analyze a received request and has the function of understanding the content of the request.

[0827] A "learning device" is a device that has the function of selecting the optimal generative AI model based on the analysis results and training that model.

[0828] A "generation device" is a device that has the function of generating specific advice in response to a user request using a selected generative AI model.

[0829] A "display device" is a device that has the function of providing and displaying the generated advice to the user.

[0830] A "natural language processing tool" is software or a library that analyzes received text data and understands and classifies its content.

[0831] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task based on a training dataset.

[0832] The present invention is a system that allows users to get reliable advice to solve specific questions and concerns about their life path. The system includes the following main components:

[0833] Data collection

[0834] The server collects quotes, personal experiences, and past advice from famous people from the internet and books. For this collection, it uses web scraping tools such as Scrapy and Beautiful Soup. The collected data is then stored in a database in text format.

[0835] Data Preparation and Training

[0836] The server then uses Hugging Face's Transformers library to organize the collected text data into a training dataset, allowing the generative AI model to learn the thought patterns and advice of celebrities. This organization process also uses natural language processing libraries such as NLTK (Natural Language Toolkit) and spaCy.

[0837] Receiving a user request

[0838] Users access the system using a dedicated app or a web browser and enter their specific concerns or questions. The text is then sent to the server via web application frameworks such as Flask and Django.

[0839] Parsing the request

[0840] The server uses spaCy to perform natural language analysis on the received request. Based on the analysis results, the request content is classified and an appropriate generative AI model is selected. The selection criterion in this case is the relevance of the request content to the training dataset.

[0841] Generating Advice

[0842] The server uses the selected generative AI model to generate specific advice based on the user's request. The generation process incorporates the thought patterns and past advice of famous people that have been trained in advance.

[0843] Validating and providing advice

[0844] The generated advice is then validated using an internal algorithm to ensure it is appropriate. Specifically, it checks whether the advice is overly biased and whether it is expressed appropriately for the context. After this validation is complete, the server sends the advice to the terminal via Flask or Django and displays it to the user.

[0845] Specific examples

[0846] Example 1: Career Choices

[0847] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0848] 2. The device receives this request and sends it to the server via Flask.

[0849] 3. The server uses spaCy to analyze the request and extract keywords such as "job change" and "workplace."

[0850] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as "Pursue your passion and take risks."

[0851] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[0852] Example 2: Academic direction

[0853] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0854] 2. The device receives this request and sends it to the server via Flask.

[0855] 3. The server uses spaCy to analyze the request and extract keywords such as "science" and "humanities."

[0856] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as, "Value your interests and curiosity, and make a choice based on which one is most suited to you."

[0857] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[0858] As described above, this system allows users to obtain reliable and specific advice, and the entire system operates through a consistent process, providing a fast and appropriate response to user requests.

[0859] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0860] Step 1:

[0861] Users access the system using a dedicated app or a web browser and enter specific concerns or questions. The entered text is sent to the server via Flask or Django.

[0862] Specific behavior: The user enters the following into the web form: "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job," and clicks the submit button. This is the input.

[0863] Step 2:

[0864] The terminal receives this request and sends it to the server as text data. The server receives the request.

[0865] Specific operation: The terminal receives input and sends a POST request to the server via Flask or Django in JSON or text format, which is the output to the server.

[0866] Step 3:

[0867] The request received by the server is analyzed. spaCy, a natural language processing tool, is used to understand the content of the request and extract keywords. This is the input. Based on the analysis results, the content of the request is classified.

[0868] Specific operation: The server analyzes a request such as "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job" using spaCy, and extracts keywords such as "job change" and "workplace." This is the output of the analysis.

[0869] Step 4:

[0870] The server selects and trains the optimal generative AI model based on the analysis results. The selection criteria are the relevance of the request content to the training dataset, which is the input. A learning device is used to select the model and conduct additional training if necessary.

[0871] Specific operation: The server selects a generative AI model for "Career Advisor" based on the analyzed keywords, and trains the model for the number of epochs using the training dataset. This is the output of model selection and training.

[0872] Step 5:

[0873] The server uses the selected generative AI model to generate specific advice in response to the user's request. The input is the user's request and the selected AI model. The generation process reflects the thinking patterns and past advice of famous people that have been trained in advance.

[0874] How it works: The server inputs a prompt such as "I'm thinking about changing jobs, but should I try a new job?" into the career advisor's generative AI model, and gets the generated advice "Follow your passion and take risks." This is the output of the generation.

[0875] Step 6:

[0876] The server validates the appropriateness of the generated advice using an internal algorithm. The input is the generated advice. It checks whether the advice is overly biased and whether it is expressed appropriately for the context.

[0877] What happens: The server validates the generated advice, "Follow your passion and take risks," to check for inappropriate content. This is the validation output:

[0878] Step 7:

[0879] If the server determines that the advice is valid, it sends it to the terminal via Flask or Django and displays it to the user. The input is the validated advice.

[0880] What happens: The server sends the advice "Follow your passion and take risks" to the terminal through the Flask and Django application and displays it on the user's screen. This is the display output:

[0881] (Application example 1)

[0882] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0883] In traditional brick-and-mortar stores, it was difficult for customers to ask questions about specific products or services in real time and receive reliable advice. There were also few ways to effectively utilize advice based on the words and personal experiences of celebrities. As a result, customers lacked the information they needed to make the best choice for themselves, leading to problems such as reduced purchasing efficiency and satisfaction.

[0884] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0885] In this invention, the server includes a means for receiving requests from users, a means for storing famous people's quotes, personal experiences, and past advice as a dataset, and a means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset. This allows customers to receive product advice in real time in a physical store. Furthermore, the selected generative AI model can be used to provide reliable advice, quickly and accurately providing customers with information to help them make the best choice for themselves.

[0886] A "terminal that receives requests from a user" is a device through which a user inputs questions or requests and transmits that information to other system components.

[0887] An "information processing device that stores the words, experiences, and past advice of famous people as a data set" is a device that collects, organizes, and stores data such as the statements and experiences of famous people.

[0888] An "information processing device that analyzes received requests and selects and trains a generative AI model based on the required training data set" is a device that analyzes requests from users, selects an AI model based on appropriate training data, and trains it.

[0889] An "information processing device that generates advice in response to a user request using a selected generative AI model" is a device that generates specific advice in response to a user question using a selected AI model.

[0890] "Means that enable customers to obtain product advice using a terminal in a physical store" refers to a system that enables customers to obtain advice on product selection and purchases through a terminal in a physical store environment.

[0891] The "terminal for providing generated advice to a user" is a device for displaying, notifying, or communicating generated advice to a user.

[0892] "Formalizing the dataset with the permission and supervision of the celebrity" means recording the statements and activities of the celebrity as official data with the celebrity's permission and supervision.

[0893] A "means for generating reliable advice" is a process or mechanism for providing accurate and useful advice to users based on reliable sources of information and data.

[0894] "Verifying the appropriateness of the generated advice using an internal algorithm" means confirming the accuracy and appropriateness of the advice generated using an internal algorithm.

[0895] "Means for adjustment as necessary" refers to a mechanism for correcting or improving the content of advice generated if it is inappropriate.

[0896] This invention is a system that allows customers to ask questions about product selection and purchases in a physical store and receive reliable advice in real time. This system consists of a terminal where the user inputs their questions, an information processing device (server) that stores and utilizes quotes and personal experiences of celebrities and past advice as a dataset, and an information processing device that trains and executes a generative AI model.

[0897] Hardware and Software

[0898] Hardware:

[0899] Smartphones, smart glasses (to input user questions and display advice)

[0900] Servers (for storing data, training and running AI models)

[0901] software:

[0902] OpenAI API (AI model generation and execution)

[0903] HTTP request library (requests)

[0904] Program processing overview

[0905] The server collects the words, experiences, and past advice of celebrities from online and offline sources and stores them as a dataset. The collected data is then used to train a generative AI model. This model learns the thought patterns and advice of celebrities and generates advice accordingly.

[0906] When a user enters a question using a smartphone or smart glasses in a physical store, the device sends the request in text format to the server. The server analyzes the request, selects the most appropriate generative AI model, and generates advice. The generated advice is verified for appropriateness by an internal algorithm, and if deemed appropriate, it is sent to the device and displayed to the user.

[0907] Specific examples

[0908] Selecting products in the store

[0909] A customer in a physical store puts on smart glasses and inputs a question such as, "Which brand is best when choosing new sneakers?"

[0910] Example prompt sentence:

[0911] Customer Question: What brand is best when choosing new sneakers?

[0912] You are an AI that provides advice based on the experiences and words of famous athletes. Please advise users on which brand of sneakers to choose by referring to the following dataset:

[0913] Dataset:

[0914] 1. Comments from famous athletes about their sneaker choices

[0915] 2. Expert-recommended brands

[0916] 3. Industry magazine rankings

[0917] This system allows customers to instantly receive helpful advice on product selection in-store and receive assistance in making the best choices.

[0918] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0919] Step 1:

[0920] The server collects quotes, personal experiences, and past advice from famous people both online and offline. The collected data is stored in text format and then compiled into training data through expert supervision. This allows the server to maintain a highly reliable data set.

[0921] Input: Text data collected from the internet and books

[0922] Output: A curated training dataset curated by experts

[0923] Specific operations: Data is obtained from online databases and books using web scraping and OCR technology, and then reviewed and tagged by experts.

[0924] Step 2:

[0925] The server trains the generative AI model using a curated training dataset, which learns the thought patterns and advice of celebrities.

[0926] Input: Training dataset

[0927] Output: A trained generative AI model

[0928] What it does: Uses a training dataset to train a model in an AI model training framework (e.g., TensorFlow, PyTorch).

[0929] Step 3:

[0930] Users use their smartphones or smart glasses in a physical store to input questions into the terminal, which then sends the questions in text format to the server.

[0931] Input: User question text

[0932] Output: Question text sent to the server

[0933] Specific operation: The user enters a question into the application's interface, which is then sent by the device to the server.

[0934] Step 4:

[0935] The server analyzes the received request and selects the most suitable generative AI model based on the relevance of the request content to the training data.

[0936] Input: User question text

[0937] Output: The selected generative AI model

[0938] Specific operation: The server uses natural language processing technology to analyze the question text and select the corresponding generative AI model.

[0939] Step 5:

[0940] The server uses the selected generative AI model to generate advice for the user's question. The generated advice is based on the thought patterns and past advice of famous people that the model has learned.

[0941] Input: User question text, generative AI model

[0942] Output: The generated advice

[0943] Specific operation: The server inputs the prompt sentence into the selected generative AI model and performs text generation using AI.

[0944] Step 6:

[0945] The server uses an internal algorithm to verify the appropriateness of the advice it generates, ensuring that it is expressed in the right context and is not overly biased.

[0946] Input: Generated advice

[0947] Output: Verified advice

[0948] What happens: The server uses internal algorithms and rule-based validation systems to check the content of the advice generated.

[0949] Step 7:

[0950] The server sends the verified advice to the user's device, which displays it to the user, who then considers his or her options and makes a decision based on the advice displayed.

[0951] Input: Verified Advice

[0952] Output: Advice displayed by the user

[0953] Specific operation: The server sends the advice in an appropriate format to the terminal, and the terminal displays the advice on the screen.

[0954] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0955] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components include a device that receives user requests, a server that stores famous quotes and personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models. The system also incorporates an emotion engine that recognizes the user's emotions and can analyze the user's emotional state when the request is made and tailor the advice accordingly.

[0956] Program processing

[0957] Data Collection and Training

[0958] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and edited by experts to create a training dataset. A generative AI model is trained using this dataset. The trained model reflects the thinking patterns and advice of famous people.

[0959] Receiving requests and analyzing sentiment

[0960] The user accesses the system and inputs specific concerns or questions in text format from a terminal. For example, they might input something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[0961] Analysis by emotion engine

[0962] The device incorporates an emotion engine that recognizes the user's emotional state when inputting a request. The emotion engine analyzes the user's emotions through analysis of the user's voice input and text, and sends the analysis results to the server. For example, if the user is feeling anxious, that emotional state will be included.

[0963] Request and emotional state analysis

[0964] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model and generates advice taking into account the user's emotional state.

[0965] AI model selection and generation

[0966] The server uses the selected generative AI model to generate advice based on the user's request and emotional state, and provides advice that matches the user's emotional state by referencing the thought patterns and past advice of famous people that the model has learned.

[0967] Validating and providing advice

[0968] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is overly biased and whether it uses appropriate language in line with the context. If necessary, it adjusts the generated advice. If the final verified and adjusted advice is deemed appropriate, it is sent to the device and displayed to the user. The user can then consider their own choices based on this advice.

[0969] Specific examples

[0970] Example 1: Career Choice and Sentiment Analysis

[0971] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[0972] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[0973] 3. The device sends the request and emotional state to the server.

[0974] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[0975] 5. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks," and adds phrases to ease anxiety.

[0976] 6. The server sends the verified advice to the terminal and displays it to the user.

[0977] Example 2: Academic orientation and sentiment analysis

[0978] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[0979] 2. The device receives this request and uses its emotion engine to analyze that the user is excited.

[0980] 3. The device sends the request and emotional state to the server.

[0981] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[0982] 5. Based on the AI ​​model, the server generates advice such as "Value your interests and curiosity, and make a choice based on which direction they are directed," and adds expressions to help the user stay calm.

[0983] 6. The server sends the verified advice to the terminal and displays it to the user.

[0984] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[0985] The processing flow will be explained below.

[0986] Step 1:

[0987] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[0988] Step 2:

[0989] The server uses the training dataset to train the generative AI model, specifically using natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[0990] Step 3:

[0991] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[0992] Step 4:

[0993] The device receives requests entered by the user and sends them in text format to the server. The device's built-in emotion engine analyzes the user's emotional state as they speak, extracting emotional states such as joy, sadness, anxiety, and excitement based on the user's voice input and text information.

[0994] Step 5:

[0995] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model. The selection criteria are the relevance of the request content to the training data and the user's emotional state.

[0996] Step 6:

[0997] The server uses the selected generative AI model to generate advice in response to the user's request. The generated advice is based on the thought patterns and past advice of famous people that the model has learned. In addition, the expression and content are adjusted according to the user's emotional state. For example, if a user is feeling anxious, reassuring expressions are used.

[0998] Step 7:

[0999] The server uses an internal algorithm to verify the appropriateness of the generated advice. It checks whether the advice is excessively biased and whether appropriate expressions are used for the context. If necessary, it makes corrections to the generated advice.

[1000] Step 8:

[1001] The server then sends the final verified and corrected advice to the terminal, which then displays it to the user.

[1002] Step 9:

[1003] Users receive advice displayed on their devices and use it to consider their own choices. For example, a user receiving advice about changing jobs can use the advice to determine the direction they truly want to take. Furthermore, advice appropriately adjusted by the emotion engine is tailored to the user's emotional state, allowing them to accept the advice with greater confidence.

[1004] In this way, the system provides users with reliable advice that takes into account their emotional state, enabling them to obtain useful information for making life choices.

[1005] Example 2

[1006] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1007] In modern society, many people are concerned about their careers, studies, and life paths. However, it is not easy to get reliable advice. In addition, there are limited advice-giving systems that take the user's emotional state into account, and general advice often lacks emotional consideration. This makes it difficult for users to make appropriate and meaningful decisions.

[1008] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a terminal that receives a request from a user, a means for saving famous quotes, personal experiences, and past advice as a dataset, a means for analyzing the received request and extracting necessary keywords and key points using natural language processing technology, a sentiment analysis means that recognizes the user's emotional state and adjusts the content of the advice based on the analysis results, a means for selecting and training a generative AI model based on a necessary training dataset, a means for generating advice in response to the user's request using the selected generative AI model, a means for verifying the appropriateness of the generated advice using an internal algorithm and adjusting it as necessary, and a terminal that provides the generated advice to the user. This makes it possible to provide highly reliable advice that takes the user's emotional state into consideration.

[1009] "User" refers to a person who accesses the system, enters their concerns and questions, and seeks advice.

[1010] "Terminal" means a device that allows a user to access the system and input / submit requests. Examples include PCs and smartphones.

[1011] A "quote" is a commonly quoted phrase or expression uttered by a famous or influential individual.

[1012] "Testimonials" refer to stories or accounts recorded by individuals about specific events or experiences.

[1013] "Past advice" refers to advice or suggestions given to others in the past.

[1014] "Dataset" refers to a set of data collected and organized for use in the System.

[1015] A "server" is a computer system that receives requests from users, processes them, and generates and provides advice.

[1016] "Natural language processing technology" refers to the techniques and algorithms that enable computers to understand and process human language.

[1017] "Keywords" refer to words or phrases that play a significant role in a user's request or text.

[1018] "Emotional state" refers to the emotions or psychological state a user experiences in a particular situation.

[1019] "Emotion analysis means" refers to the technology or algorithm that recognizes emotions from user requests or voice and provides the analysis results.

[1020] A "generative AI model" is an artificial intelligence model that is trained using a large dataset and is used to generate advice in response to a user request.

[1021] "Relevance" refers to the unbiased nature of the advice generated, and whether it is natural, contextual, and appropriate.

[1022] "Internal algorithm" refers to the calculation methods and programs used within the system.

[1023] "Validation" refers to the process of evaluating whether the generated advice is appropriate.

[1024] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores quotes, personal experiences, and past advice as a dataset, and a server that trains and executes a generative AI model. The system also incorporates an emotion analysis means that recognizes the user's emotions, and can analyze the user's emotional state when the request is made and adjust the content of the advice accordingly.

[1025] The server uses Python scripts and web analytics tools to collect quotes and testimonials from online sources and books, and store them in a database in text format. Specifically, the Beautiful Soup and Requests libraries are used to scrape data from websites, the Pandas library is used to format and clean the data, and then the data is stored in a MySQL database via SQLAlchemy.

[1026] The collected dataset is used to train a generative AI model (e.g., GPT-3) using GPUs on cloud computing platforms (e.g., AWS or Google Cloud). The training process is carried out over hundreds of epochs using libraries such as TensorFlow and PyTorch.

[1027] Users access the web application using a device such as a PC or smartphone and enter specific concerns or questions in text format. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The device then asynchronously sends this request to the server using JavaScript (AJAX).

[1028] The device uses emotion analysis to recognize the user's emotional state from the input text. For example, it uses the BERT model with the Hugging Face transformers library to classify emotions and obtain emotion labels such as "anxiety" or "excitement." The analysis results are sent to the server in JSON format.

[1029] The server combines the request and the emotional state and uses natural language processing (NLP) techniques to analyze the request. For example, it uses the spaCy library to tokenize the text, extract nouns and verbs, and identify keywords. Keywords such as "job change," "new workplace," and "challenge" are extracted.

[1030] The server selects the most suitable generative AI model based on the analysis results. It uses a keyword matching algorithm to select the best one from multiple celebrity models. If the keyword "challenge" is included, it selects a model that provides adventurous advice.

[1031] Using the selected generative AI model, the server generates advice based on the user's request and emotional state. For example, using GPT-3, the server generates advice such as, "Identify your passion and have the courage to take risks." The prompt used is, "I'm thinking about changing jobs. I'm feeling anxious. What should I do?"

[1032] The server uses internal algorithms to verify the appropriateness of the generated advice and adjusts it as needed, for example by using rule-based filters to ensure that it does not contain extreme or inappropriate language. Finally, advice that passes the verification is sent to the device and displayed to the user.

[1033] Specific examples of prompts are as follows:

[1034] "I'm thinking about changing jobs, but I'm not sure whether I should stay where I am or try something new. What should I do? I'm feeling anxious."

[1035] "I'm worried about my further education. I'm not sure whether to go into science or humanities, but I'm feeling excited. I'd appreciate some advice."

[1036] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[1037] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1038] Step 1: User request input

[1039] A user enters specific questions or concerns about their life path into a text box in a web application. For example, they might enter, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job." This is the input data. The device asynchronously sends the text entered by the user to the server using AJAX. This sent data is the input for the next step.

[1040] Step 2: Receiving the request

[1041] The terminal receives text input from the user and passes it to the emotion analysis means. The input data is the text information sent by the user. The server receives the text received via an AJAX request for analysis. The output is the user's input text itself.

[1042] Step 3: Analyze emotional state

[1043] The device passes the received text to the sentiment analyzer. The sentiment analyzer uses, for example, the Hugging Face transformers library and analyzes the text with the BERT model. The input is the user's text received in step 2. The BERT model tokenizes the text and infers an emotion label based on the context. The analysis results indicate that the user is feeling anxious. The analysis results are sent to the server in JSON format.

[1044] Step 4: Parsing the request

[1045] The server analyzes the sentiment analysis results received in JSON format and the request text using natural language processing technology such as SpaCy. The input is the user's text and the sentiment analysis results. The text is tokenized using the SpaCy library, and key keywords such as "job change," "challenge," and "current workplace" are extracted. The analysis results become the input data for the next step.

[1046] Step 5: Selecting a generative AI model

[1047] The server selects the most suitable generative AI model based on the request content and the results of the sentiment analysis. The inputs are the keywords and emotional state extracted in step 4. A keyword matching algorithm is used to calculate the degree of match with the tags of each celebrity model and select the most suitable celebrity model. For example, based on the keywords "challenge" and "job change," a model that provides adventurous advice is selected. The results of this selection are the input data for the next step.

[1048] Step 6: Generating Advice

[1049] The server uses the selected generative AI model to generate advice appropriate to the user's request and emotional state. The input is the selected generative AI model and a prompt. Specifically, the prompt "I'm thinking about changing jobs. I'm feeling anxious. What should I think about this?" is input into GPT-3, and the server generates the advice "Identify where your passion lies and have the courage to take on challenges without fear of risk." This is the output data.

[1050] Step 7: Verify the advice

[1051] The server uses an internal algorithm to verify the appropriateness of the generated advice. The input is the advice generated in step 6. A rule-based filter is used to check whether it contains extreme expressions or inappropriate sentences. For example, regular expressions are used to check whether specific keywords are included and whether the expression is natural in the context. Adjustments are made as necessary, and finally, advice that passes the verification is generated as output data.

[1052] Step 8: Providing advice

[1053] The server finally sends the advice that has passed verification to the terminal and displays it to the user. The input is verified advice. The terminal uses HTML and CSS to display the received advice in an easy-to-read GUI (Graphical User Interface). Important points are highlighted to make it easier for the user to understand the advice.

[1054] (Application example 2)

[1055] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1056] Currently, there are limited ways to quickly provide appropriate advice to employees in brick-and-mortar stores regarding work-related or career-related issues. It is also difficult to provide advice that takes into account the emotional state of employees, resulting in problems such as a decline in employee motivation and the accumulation of stress. Furthermore, there is no established method for ensuring the reliability of the advice provided.

[1057] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving requests from users, means for saving famous people's quotes, personal experiences, and past advice as a dataset, means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset, means for generating advice in response to the user request using the selected generative AI model, an emotion engine for analyzing the user's emotional state, and an application for solving problems between employees and providing feedback. This makes it possible to quickly provide reliable advice that takes into account the emotional state of employees.

[1058] A "terminal" is a device that receives requests from users and provides generated advice.

[1059] "Server" means a central processing unit for storing and analyzing data, selecting and training generative AI models, and generating and providing advice.

[1060] A "celebrity" refers to a person who is widely known in society, and whose famous quotes, experiences, advice, etc. are used as a dataset.

[1061] A "dataset" is a collection of famous quotes, personal experiences, and past advice that is used as training material for generative AI models.

[1062] A "generative AI model" is an artificial intelligence model that generates advice in response to a user request based on a training dataset.

[1063] The "emotion engine" is software that analyzes the user's emotional state when they input a request and adjusts the advice content based on the results.

[1064] The "application for problem-solving and feedback among employees" is software that provides advice to employees in physical stores, taking into account their emotional state, in response to their concerns and questions.

[1065] "Internal Algorithm" means an internal computational means for validating the appropriateness of the generated advice and adjusting it if necessary.

[1066] A system for implementing the present invention includes the following elements:

[1067] Hardware and Software

[1068] Hardware: Smartphone

[1069] software:

[1070] Database: MySQL, PostgreSQL, etc.

[1071] Emotion engine: EmotionAPI

[1072] Generative AI model: OpenAI GPT-4

[1073] Application backend: Node.js

[1074] Frontend: React Native

[1075] Explanation of program processing

[1076] First, the server collects quotes, personal experiences, and past advice from famous people online and from books. This data is stored in a text database (MySQL or PostgreSQL) and cleaned under the supervision of experts. The cleaned data is organized into a training dataset, and a generative AI model (OpenAI GPT-4) is trained using this data.

[1077] Users access the application with specific concerns or questions and enter their request in text form via their smartphone. For example, they can enter something like, "I'm not sure whether I should apply for a new position or stay in my current one."

[1078] The device is equipped with an emotion engine (Emotion API) that analyzes the user's emotional state when they input a request. The emotion engine detects the user's emotions from the text input and sends the emotion analysis results to the server.

[1079] The server analyzes the received request and the results of sentiment analysis, and selects the most suitable generative AI model. This analysis is performed using syntactic analysis technology to extract keywords and important points contained in the request.

[1080] The server then uses the selected generative AI model to generate advice. The generative AI model references the learned thinking patterns and past advice of famous people to provide advice tailored to the user's concerns and emotional state. OpenAI GPT-4 is used to generate this advice.

[1081] The generated advice is then validated by an internal algorithm to ensure its appropriateness, including whether the advice is overly biased or if the wording is appropriate. If necessary, the advice is adjusted.

[1082] The final verified and adjusted advice is sent back to the terminal and provided to the user, who can then consider their own choices based on this advice.

[1083] Specific examples

[1084] Example 1: Career Choice and Sentiment Analysis

[1085] 1. The user enters the following text: "I'm not sure whether I should apply for a new position or stay in my current one."

[1086] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[1087] 3. The device sends the request and emotional state to the server.

[1088] 4. The server selects an appropriate generative AI model based on the request and the results of sentiment analysis.

[1089] 5. The server generates advice based on the generated AI model, such as "Identify your passion and have the courage to take on challenges without fear of risk," and adds expressions to ease anxiety.

[1090] 6. The server sends the verified advice to the terminal and displays it to the user.

[1091] Prompt Sentence Examples

[1092] "I'm not sure whether I should apply for a new position or stay in my current one. How should I decide? Helplessness: I feel it. Anxiety: I feel it. Confidence: I want it."

[1093] In this way, the system provides reliable advice that takes into account the user's emotional state, enabling store staff to solve problems and provide feedback.

[1094] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1095] Step 1:

[1096] Famous quotes, personal experiences, and past advice from famous people are collected online and from books. This data is stored in text format in a database (MySQL or PostgreSQL). The collected data is then cleaned and reviewed by experts and organized into a training dataset. This ensures high-quality data is used to train generative AI models.

[1097] Input: Text data collected from online sources and books

[1098] Output: Data organized as a training dataset

[1099] Step 2:

[1100] The prepared training dataset will be used to train a generative AI model (OpenAI GPT-4), which will then learn the thought patterns and advice of celebrities, laying the foundation for generating appropriate advice in response to user requests.

[1101] Input: A curated training dataset

[1102] Output: A trained generative AI model

[1103] Step 3:

[1104] The user enters specific concerns or questions in text format via their smartphone. For example, they might enter something like, "I'm not sure whether I should apply for a new position or stay in my current one." This text data is then analyzed in the next step.

[1105] Input: The textual request entered by the user

[1106] Output: Request data in text format

[1107] Step 4:

[1108] The emotion engine (Emotion API) built into the device analyzes the user's emotional state when they input a request. The emotion engine analyzes the text data and detects the emotion the user is feeling (e.g., anxiety or helplessness). The detected emotional state is sent to the server.

[1109] Input: Request data in text format

[1110] Output: Emotional analysis results (e.g. anxiety, helplessness)

[1111] Step 5:

[1112] The server analyzes the received request and the sentiment analysis results, and performs a syntactic analysis to extract keywords and key points contained in the request. This analysis clarifies the important points of the request content.

[1113] Input: Request data in text format, sentiment analysis results

[1114] Output: Keywords and key points of the request

[1115] Step 6:

[1116] The server selects the most suitable generative AI model based on the extracted keywords and key points, and uses the selected generative AI model to generate advice in response to the user's request. The generated advice also takes into account the user's emotional state.

[1117] Input: Keywords and key points of the request, sentiment analysis results

[1118] Output: Generated advice

[1119] Step 7:

[1120] The generated advice is verified by an algorithm inside the server to check its appropriateness, checking whether the advice is excessively biased and whether appropriate wording is used, and adjusting the advice content as necessary.

[1121] Input: Generated advice

[1122] Output: Validated and adjusted advice

[1123] Step 8:

[1124] The final verified and adjusted advice is then sent back to the device and provided to the user, who can then receive the advice via their smartphone and consider their own choices.

[1125] Input: Validated and adjusted advice

[1126] Output: Advice given to the user

[1127] The above steps make it possible to provide fast, reliable advice that takes into account the emotional state of employees in physical stores in response to their concerns and questions.

[1128] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1129] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1130] 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 the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1131] [Fourth embodiment]

[1132] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1133] 7, a 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.

[1134] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1135] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1136] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1137] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1138] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1139] The control object 443 includes a display device, LEDs in the eyes, and motors for driving 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 emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1140] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1141] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1142] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1143] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1144] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1145] The present invention is a system that enables users to obtain reliable advice to solve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores famous quotes, personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models.

[1146] Program processing

[1147] Data Collection and Training

[1148] The server collects quotes, personal experiences, and past advice from famous people online and in books. The collected data is saved in text format and then edited by experts to be ready for use as a training dataset. This dataset is used to train a generative AI model. The trained model reflects the thinking patterns and advice of famous people.

[1149] Receiving and parsing the request

[1150] The user accesses the system and inputs specific concerns or questions into the terminal. For example, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[1151] AI model selection and generation

[1152] The server analyzes the received request and selects the celebrity generative AI model that best suits the request. The selection criteria are the relevance between the request content and the training data. The selected generative AI model is used to generate advice that meets the user's request. The generation process ensures that reliable advice is provided based on the celebrity's thought patterns and past advice that the model has learned.

[1153] Providing advice

[1154] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is not overly biased and whether it is expressed in an appropriate context. If the verified advice is deemed appropriate, it is sent to the terminal and displayed to the user. The user can then consider their own choices and make a decision based on this advice.

[1155] Specific examples

[1156] Example 1: Career Choices

[1157] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[1158] 2. The device receives this request and sends it to the server.

[1159] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model with extensive experience in career decisions.

[1160] 4. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks."

[1161] 5. The server sends the verified advice to the terminal and displays it to the user.

[1162] Example 2: Academic direction

[1163] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[1164] 2. The device receives this request and sends it to the server.

[1165] 3. The server analyzes the request and selects an appropriate celebrity generative AI model, for example, a celebrity model that is good at giving academic advice.

[1166] 4. Based on the AI ​​model, the server generates advice such as, "Value your own interests and curiosity, and make a choice based on which one is best suited to you."

[1167] 5. The server sends the verified advice to the terminal and displays it to the user.

[1168] This system allows users to obtain reliable advice that can be used effectively in making life choices.

[1169] The processing flow will be explained below.

[1170] Step 1:

[1171] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[1172] Step 2:

[1173] The server uses the training dataset to train the generative AI model. Specifically, it uses natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[1174] Step 3:

[1175] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[1176] Step 4:

[1177] The terminal receives a request input by the user and transmits the content in text format to the server, which then formats the request appropriately for the system.

[1178] Step 5:

[1179] The server analyzes the received request. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable AI model to generate a celebrity.

[1180] Step 6:

[1181] The server uses the selected generative AI model to generate advice for the user's request, referencing the thinking patterns and past advice of famous people that the model has learned, and generates advice that is specifically suited to the request.

[1182] Step 7:

[1183] The server uses an internal algorithm to verify the appropriateness of the advice it generates, for example, by checking that the advice is not overly biased and that it uses appropriate language in the context. If necessary, it adjusts the advice it generates.

[1184] Step 8:

[1185] The server then sends the final verified and adjusted advice to the terminal, where it is displayed to the user.

[1186] Step 9:

[1187] Users receive advice displayed on their devices and use it to consider their choices, helping them make better decisions about their life paths.

[1188] In this way, the system provides users with reliable advice and allows them to obtain useful information in making life choices.

[1189] Example 1

[1190] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1191] In today's world, when individuals face various life choices, it is difficult to obtain advice based on reliable information. Furthermore, existing advice systems are general and cannot provide specific advice tailored to individual concerns or questions. Therefore, there is a need for the development of a system that can address the specific problems faced by individual users and provide accurate advice based on deep insight and experience.

[1192] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1193] In this invention, the server includes an information processing device that receives requests from users, a data storage device that uses a database to store quotes and personal experiences, an analysis device that analyzes the received requests using a natural language processing tool, a learning device that selects and trains an optimal generative AI model based on the analysis results, a generation device that generates advice in response to the user request using the selected generative AI model, and a display device that provides the generated advice to the user. This makes it possible to provide highly reliable and specific advice based on the concerns and questions of individual users.

[1194] An "information processing device" is a device capable of receiving input from a user and forwarding it to other system components.

[1195] A "data storage device" is a device that stores data such as famous quotes and personal experiences in a database and has the function of retrieving that data as needed.

[1196] An "analysis device" is a device that uses natural language processing tools to analyze a received request and has the function of understanding the content of the request.

[1197] A "learning device" is a device that has the function of selecting the optimal generative AI model based on the analysis results and training that model.

[1198] A "generation device" is a device that has the function of generating specific advice in response to a user request using a selected generative AI model.

[1199] A "display device" is a device that has the function of providing and displaying the generated advice to the user.

[1200] A "natural language processing tool" is software or a library that analyzes received text data and understands and classifies its content.

[1201] A "generative AI model" is an artificial intelligence model that is trained to perform a specific task based on a training dataset.

[1202] The present invention is a system that allows users to get reliable advice to solve specific questions and concerns about their life path. The system includes the following main components:

[1203] Data collection

[1204] The server collects quotes, personal experiences, and past advice from famous people from the internet and books. For this collection, it uses web scraping tools such as Scrapy and Beautiful Soup. The collected data is then stored in a database in text format.

[1205] Data Preparation and Training

[1206] The server then uses Hugging Face's Transformers library to organize the collected text data into a training dataset, allowing the generative AI model to learn the thought patterns and advice of celebrities. This organization process also uses natural language processing libraries such as NLTK (Natural Language Toolkit) and spaCy.

[1207] Receiving a user request

[1208] Users access the system using a dedicated app or a web browser and enter their specific concerns or questions. The text is then sent to the server via web application frameworks such as Flask and Django.

[1209] Parsing the request

[1210] The server uses spaCy to perform natural language analysis on the received request. Based on the analysis results, the request content is classified and an appropriate generative AI model is selected. The selection criterion in this case is the relevance of the request content to the training dataset.

[1211] Generating Advice

[1212] The server uses the selected generative AI model to generate specific advice based on the user's request. The generation process incorporates the thought patterns and past advice of famous people that have been trained in advance.

[1213] Validating and providing advice

[1214] The generated advice is then validated using an internal algorithm to ensure it is appropriate. Specifically, it checks whether the advice is overly biased and whether it is expressed appropriately for the context. After this validation is complete, the server sends the advice to the terminal via Flask or Django and displays it to the user.

[1215] Specific examples

[1216] Example 1: Career Choices

[1217] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[1218] 2. The device receives this request and sends it to the server via Flask.

[1219] 3. The server uses spaCy to analyze the request and extract keywords such as "job change" and "workplace."

[1220] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as "Pursue your passion and take risks."

[1221] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[1222] Example 2: Academic direction

[1223] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[1224] 2. The device receives this request and sends it to the server via Flask.

[1225] 3. The server uses spaCy to analyze the request and extract keywords such as "science" and "humanities."

[1226] 4. The server selects an appropriate generative AI model and uses the Transformers library to generate advice such as, "Value your interests and curiosity, and make a choice based on which one is most suited to you."

[1227] 5. The server validates this advice using its internal algorithm, and if it is deemed appropriate, sends it to the terminal via Flask and displays it to the user.

[1228] As described above, this system allows users to obtain reliable and specific advice, and the entire system operates through a consistent process, providing a fast and appropriate response to user requests.

[1229] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1230] Step 1:

[1231] Users access the system using a dedicated app or a web browser and enter specific concerns or questions. The entered text is sent to the server via Flask or Django.

[1232] Specific behavior: The user enters the following into the web form: "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job," and clicks the submit button. This is the input.

[1233] Step 2:

[1234] The terminal receives this request and sends it to the server as text data. The server receives the request.

[1235] Specific operation: The terminal receives input and sends a POST request to the server via Flask or Django in JSON or text format, which is the output to the server.

[1236] Step 3:

[1237] The request received by the server is analyzed. spaCy, a natural language processing tool, is used to understand the content of the request and extract keywords. This is the input. Based on the analysis results, the content of the request is classified.

[1238] Specific operation: The server analyzes a request such as "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job" using spaCy, and extracts keywords such as "job change" and "workplace." This is the output of the analysis.

[1239] Step 4:

[1240] The server selects and trains the optimal generative AI model based on the analysis results. The selection criteria are the relevance of the request content to the training dataset, which is the input. A learning device is used to select the model and conduct additional training if necessary.

[1241] Specific operation: The server selects a generative AI model for "Career Advisor" based on the analyzed keywords, and trains the model for the number of epochs using the training dataset. This is the output of model selection and training.

[1242] Step 5:

[1243] The server uses the selected generative AI model to generate specific advice in response to the user's request. The input is the user's request and the selected AI model. The generation process reflects the thinking patterns and past advice of famous people that have been trained in advance.

[1244] How it works: The server inputs a prompt such as "I'm thinking about changing jobs, but should I try a new job?" into the career advisor's generative AI model, and gets the generated advice "Follow your passion and take risks." This is the output of the generation.

[1245] Step 6:

[1246] The server validates the appropriateness of the generated advice using an internal algorithm. The input is the generated advice. It checks whether the advice is overly biased and whether it is expressed appropriately for the context.

[1247] What happens: The server validates the generated advice, "Follow your passion and take risks," to check for inappropriate content. This is the validation output:

[1248] Step 7:

[1249] If the server determines that the advice is valid, it sends it to the terminal via Flask or Django and displays it to the user. The input is the validated advice.

[1250] What happens: The server sends the advice "Follow your passion and take risks" to the terminal through the Flask and Django application and displays it on the user's screen. This is the display output:

[1251] (Application example 1)

[1252] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1253] In traditional brick-and-mortar stores, it was difficult for customers to ask questions about specific products or services in real time and receive reliable advice. There were also few ways to effectively utilize advice based on the words and personal experiences of celebrities. As a result, customers lacked the information they needed to make the best choice for themselves, leading to problems such as reduced purchasing efficiency and satisfaction.

[1254] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1255] In this invention, the server includes a means for receiving requests from users, a means for storing famous people's quotes, personal experiences, and past advice as a dataset, and a means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset. This allows customers to receive product advice in real time in a physical store. Furthermore, the selected generative AI model can be used to provide reliable advice, quickly and accurately providing customers with information to help them make the best choice for themselves.

[1256] A "terminal that receives requests from a user" is a device through which a user inputs questions or requests and transmits that information to other system components.

[1257] An "information processing device that stores the words, experiences, and past advice of famous people as a data set" is a device that collects, organizes, and stores data such as the statements and experiences of famous people.

[1258] An "information processing device that analyzes received requests and selects and trains a generative AI model based on the required training data set" is a device that analyzes requests from users, selects an AI model based on appropriate training data, and trains it.

[1259] An "information processing device that generates advice in response to a user request using a selected generative AI model" is a device that generates specific advice in response to a user question using a selected AI model.

[1260] "Means that enable customers to obtain product advice using a terminal in a physical store" refers to a system that enables customers to obtain advice on product selection and purchases through a terminal in a physical store environment.

[1261] The "terminal for providing generated advice to a user" is a device for displaying, notifying, or communicating generated advice to a user.

[1262] "Formalizing the dataset with the permission and supervision of the celebrity" means recording the statements and activities of the celebrity as official data with the celebrity's permission and supervision.

[1263] A "means for generating reliable advice" is a process or mechanism for providing accurate and useful advice to users based on reliable sources of information and data.

[1264] "Verifying the appropriateness of the generated advice using an internal algorithm" means confirming the accuracy and appropriateness of the advice generated using an internal algorithm.

[1265] "Means for adjustment as necessary" refers to a mechanism for correcting or improving the content of advice generated if it is inappropriate.

[1266] This invention is a system that allows customers to ask questions about product selection and purchases in a physical store and receive reliable advice in real time. This system consists of a terminal where the user inputs their questions, an information processing device (server) that stores and utilizes quotes and personal experiences of celebrities and past advice as a dataset, and an information processing device that trains and executes a generative AI model.

[1267] Hardware and Software

[1268] Hardware:

[1269] Smartphones, smart glasses (to input user questions and display advice)

[1270] Servers (for storing data, training and running AI models)

[1271] software:

[1272] OpenAI API (AI model generation and execution)

[1273] HTTP request library (requests)

[1274] Program processing overview

[1275] The server collects the words, experiences, and past advice of celebrities from online and offline sources and stores them as a dataset. The collected data is then used to train a generative AI model. This model learns the thought patterns and advice of celebrities and generates advice accordingly.

[1276] When a user enters a question using a smartphone or smart glasses in a physical store, the device sends the request in text format to the server. The server analyzes the request, selects the most appropriate generative AI model, and generates advice. The generated advice is verified for appropriateness by an internal algorithm, and if deemed appropriate, it is sent to the device and displayed to the user.

[1277] Specific examples

[1278] Selecting products in the store

[1279] A customer in a physical store puts on smart glasses and inputs a question such as, "Which brand is best when choosing new sneakers?"

[1280] Example prompt sentence:

[1281] Customer Question: What brand is best when choosing new sneakers?

[1282] You are an AI that provides advice based on the experiences and words of famous athletes. Please advise users on which brand of sneakers to choose by referring to the following dataset:

[1283] Dataset:

[1284] 1. Comments from famous athletes about their sneaker choices

[1285] 2. Expert-recommended brands

[1286] 3. Industry magazine rankings

[1287] This system allows customers to instantly receive helpful advice on product selection in-store and receive assistance in making the best choices.

[1288] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1289] Step 1:

[1290] The server collects quotes, personal experiences, and past advice from famous people both online and offline. The collected data is stored in text format and then compiled into training data through expert supervision. This allows the server to maintain a highly reliable data set.

[1291] Input: Text data collected from the internet and books

[1292] Output: A curated training dataset curated by experts

[1293] Specific operations: Data is obtained from online databases and books using web scraping and OCR technology, and then reviewed and tagged by experts.

[1294] Step 2:

[1295] The server trains the generative AI model using a curated training dataset, which learns the thought patterns and advice of celebrities.

[1296] Input: Training dataset

[1297] Output: A trained generative AI model

[1298] What it does: Uses a training dataset to train a model in an AI model training framework (e.g., TensorFlow, PyTorch).

[1299] Step 3:

[1300] Users use their smartphones or smart glasses in a physical store to input questions into the terminal, which then sends the questions in text format to the server.

[1301] Input: User question text

[1302] Output: Question text sent to the server

[1303] Specific operation: The user enters a question into the application's interface, which is then sent by the device to the server.

[1304] Step 4:

[1305] The server analyzes the received request and selects the most suitable generative AI model based on the relevance of the request content to the training data.

[1306] Input: User question text

[1307] Output: The selected generative AI model

[1308] Specific operation: The server uses natural language processing technology to analyze the question text and select the corresponding generative AI model.

[1309] Step 5:

[1310] The server uses the selected generative AI model to generate advice for the user's question. The generated advice is based on the thought patterns and past advice of famous people that the model has learned.

[1311] Input: User question text, generative AI model

[1312] Output: The generated advice

[1313] Specific operation: The server inputs the prompt sentence into the selected generative AI model and performs text generation using AI.

[1314] Step 6:

[1315] The server uses an internal algorithm to verify the appropriateness of the advice it generates, ensuring that it is expressed in the right context and is not overly biased.

[1316] Input: Generated advice

[1317] Output: Verified advice

[1318] What happens: The server uses internal algorithms and rule-based validation systems to check the content of the advice generated.

[1319] Step 7:

[1320] The server sends the verified advice to the user's device, which displays it to the user, who then considers his or her options and makes a decision based on the advice displayed.

[1321] Input: Verified Advice

[1322] Output: Advice displayed by the user

[1323] Specific operation: The server sends the advice in an appropriate format to the terminal, and the terminal displays the advice on the screen.

[1324] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1325] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components include a device that receives user requests, a server that stores famous quotes and personal experiences, and past advice as a dataset, and a server that trains and executes generative AI models. The system also incorporates an emotion engine that recognizes the user's emotions and can analyze the user's emotional state when the request is made and tailor the advice accordingly.

[1326] Program processing

[1327] Data Collection and Training

[1328] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and edited by experts to create a training dataset. A generative AI model is trained using this dataset. The trained model reflects the thinking patterns and advice of famous people.

[1329] Receiving requests and analyzing sentiment

[1330] The user accesses the system and inputs specific concerns or questions in text format from a terminal. For example, they might input something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The terminal then sends this request in text format to the server.

[1331] Analysis by emotion engine

[1332] The device incorporates an emotion engine that recognizes the user's emotional state when inputting a request. The emotion engine analyzes the user's emotions through analysis of the user's voice input and text, and sends the analysis results to the server. For example, if the user is feeling anxious, that emotional state will be included.

[1333] Request and emotional state analysis

[1334] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model and generates advice taking into account the user's emotional state.

[1335] AI model selection and generation

[1336] The server uses the selected generative AI model to generate advice based on the user's request and emotional state, and provides advice that matches the user's emotional state by referencing the thought patterns and past advice of famous people that the model has learned.

[1337] Validating and providing advice

[1338] The server uses an internal algorithm to verify the appropriateness of the generated advice. For example, it checks whether the advice is overly biased and whether it uses appropriate language in line with the context. If necessary, it adjusts the generated advice. If the final verified and adjusted advice is deemed appropriate, it is sent to the device and displayed to the user. The user can then consider their own choices based on this advice.

[1339] Specific examples

[1340] Example 1: Career Choice and Sentiment Analysis

[1341] 1. The user types, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay where I am."

[1342] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[1343] 3. The device sends the request and emotional state to the server.

[1344] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[1345] 5. Based on the AI ​​model, the server generates advice such as, "Identify your passion and have the courage to take risks," and adds phrases to ease anxiety.

[1346] 6. The server sends the verified advice to the terminal and displays it to the user.

[1347] Example 2: Academic orientation and sentiment analysis

[1348] 1. The user enters, "I'm not sure whether to pursue science or humanities."

[1349] 2. The device receives this request and uses its emotion engine to analyze that the user is excited.

[1350] 3. The device sends the request and emotional state to the server.

[1351] 4. The server selects an appropriate celebrity generation AI model based on the request and the results of sentiment analysis.

[1352] 5. Based on the AI ​​model, the server generates advice such as "Value your interests and curiosity, and make a choice based on which direction they are directed," and adds expressions to help the user stay calm.

[1353] 6. The server sends the verified advice to the terminal and displays it to the user.

[1354] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[1355] The processing flow will be explained below.

[1356] Step 1:

[1357] The server collects quotes, personal experiences, and past advice from famous people online and from books. The collected data is stored in a database in text format. The stored data is then cleaned and compiled into a training dataset through expert supervision.

[1358] Step 2:

[1359] The server uses the training dataset to train the generative AI model, specifically using natural language processing (NLP) algorithms to learn the thought patterns and advice of celebrities, enabling it to generate highly accurate advice.

[1360] Step 3:

[1361] Users access the system and enter specific concerns or questions in text format on their terminal. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one."

[1362] Step 4:

[1363] The device receives requests entered by the user and sends them in text format to the server. The device's built-in emotion engine analyzes the user's emotional state as they speak, extracting emotional states such as joy, sadness, anxiety, and excitement based on the user's voice input and text information.

[1364] Step 5:

[1365] The server analyzes the received request and the emotional state from the emotion engine. Specifically, it parses the request content and extracts keywords and key points contained in the request. Based on the results of this analysis, it selects the most suitable celebrity generation AI model. The selection criteria are the relevance of the request content to the training data and the user's emotional state.

[1366] Step 6:

[1367] The server uses the selected generative AI model to generate advice in response to the user's request. The generated advice is based on the thought patterns and past advice of famous people that the model has learned. In addition, the expression and content are adjusted according to the user's emotional state. For example, if a user is feeling anxious, reassuring expressions are used.

[1368] Step 7:

[1369] The server uses an internal algorithm to verify the appropriateness of the generated advice. It checks whether the advice is excessively biased and whether appropriate expressions are used for the context. If necessary, it makes corrections to the generated advice.

[1370] Step 8:

[1371] The server then sends the final verified and corrected advice to the terminal, which then displays it to the user.

[1372] Step 9:

[1373] Users receive advice displayed on their devices and use it to consider their own choices. For example, a user receiving advice about changing jobs can use the advice to determine the direction they truly want to take. Furthermore, advice appropriately adjusted by the emotion engine is tailored to the user's emotional state, allowing them to accept the advice with greater confidence.

[1374] In this way, the system provides users with reliable advice that takes into account their emotional state, enabling them to obtain useful information for making life choices.

[1375] Example 2

[1376] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1377] In modern society, many people are concerned about their careers, studies, and life paths. However, it is not easy to get reliable advice. In addition, there are limited advice-giving systems that take the user's emotional state into account, and general advice often lacks emotional consideration. This makes it difficult for users to make appropriate and meaningful decisions.

[1378] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a terminal that receives a request from a user, a means for saving famous quotes, personal experiences, and past advice as a dataset, a means for analyzing the received request and extracting necessary keywords and key points using natural language processing technology, a sentiment analysis means that recognizes the user's emotional state and adjusts the content of the advice based on the analysis results, a means for selecting and training a generative AI model based on a necessary training dataset, a means for generating advice in response to the user's request using the selected generative AI model, a means for verifying the appropriateness of the generated advice using an internal algorithm and adjusting it as necessary, and a terminal that provides the generated advice to the user. This makes it possible to provide highly reliable advice that takes the user's emotional state into consideration.

[1379] "User" refers to a person who accesses the system, enters their concerns and questions, and seeks advice.

[1380] "Terminal" means a device that allows a user to access the system and input / submit requests. Examples include PCs and smartphones.

[1381] A "quote" is a commonly quoted phrase or expression uttered by a famous or influential individual.

[1382] "Testimonials" refer to stories or accounts recorded by individuals about specific events or experiences.

[1383] "Past advice" refers to advice or suggestions given to others in the past.

[1384] "Dataset" refers to a set of data collected and organized for use in the System.

[1385] A "server" is a computer system that receives requests from users, processes them, and generates and provides advice.

[1386] "Natural language processing technology" refers to the techniques and algorithms that enable computers to understand and process human language.

[1387] "Keywords" refer to words or phrases that play a significant role in a user's request or text.

[1388] "Emotional state" refers to the emotions or psychological state a user experiences in a particular situation.

[1389] "Emotion analysis means" refers to the technology or algorithm that recognizes emotions from user requests or voice and provides the analysis results.

[1390] A "generative AI model" is an artificial intelligence model that is trained using a large dataset and is used to generate advice in response to a user request.

[1391] "Relevance" refers to the unbiased nature of the advice generated, and whether it is natural, contextual, and appropriate.

[1392] "Internal algorithm" refers to the calculation methods and programs used within the system.

[1393] "Validation" refers to the process of evaluating whether the generated advice is appropriate.

[1394] The present invention is a system that enables users to receive reliable advice to resolve specific questions and concerns about their life path. The system's main components are a terminal that receives requests from users, a server that stores quotes, personal experiences, and past advice as a dataset, and a server that trains and executes a generative AI model. The system also incorporates an emotion analysis means that recognizes the user's emotions, and can analyze the user's emotional state when the request is made and adjust the content of the advice accordingly.

[1395] The server uses Python scripts and web analytics tools to collect quotes and testimonials from online sources and books, and store them in a database in text format. Specifically, the Beautiful Soup and Requests libraries are used to scrape data from websites, the Pandas library is used to format and clean the data, and then the data is stored in a MySQL database via SQLAlchemy.

[1396] The collected dataset is used to train a generative AI model (e.g., GPT-3) using GPUs on cloud computing platforms (e.g., AWS or Google Cloud). The training process is carried out over hundreds of epochs using libraries such as TensorFlow and PyTorch.

[1397] Users access the web application using a device such as a PC or smartphone and enter specific concerns or questions in text format. For example, they might enter something like, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current one." The device then asynchronously sends this request to the server using JavaScript (AJAX).

[1398] The device uses emotion analysis to recognize the user's emotional state from the input text. For example, it uses the BERT model with the Hugging Face transformers library to classify emotions and obtain emotion labels such as "anxiety" or "excitement." The analysis results are sent to the server in JSON format.

[1399] The server combines the request and the emotional state and uses natural language processing (NLP) techniques to analyze the request. For example, it uses the spaCy library to tokenize the text, extract nouns and verbs, and identify keywords. Keywords such as "job change," "new workplace," and "challenge" are extracted.

[1400] The server selects the most suitable generative AI model based on the analysis results. It uses a keyword matching algorithm to select the best one from multiple celebrity models. If the keyword "challenge" is included, it selects a model that provides adventurous advice.

[1401] Using the selected generative AI model, the server generates advice based on the user's request and emotional state. For example, using GPT-3, the server generates advice such as, "Identify your passion and have the courage to take risks." The prompt used is, "I'm thinking about changing jobs. I'm feeling anxious. What should I do?"

[1402] The server uses internal algorithms to verify the appropriateness of the generated advice and adjusts it as needed, for example by using rule-based filters to ensure that it does not contain extreme or inappropriate language. Finally, advice that passes the verification is sent to the device and displayed to the user.

[1403] Specific examples of prompts are as follows:

[1404] "I'm thinking about changing jobs, but I'm not sure whether I should stay where I am or try something new. What should I do? I'm feeling anxious."

[1405] "I'm worried about my further education. I'm not sure whether to go into science or humanities, but I'm feeling excited. I'd appreciate some advice."

[1406] In this way, the system provides reliable advice that takes into account the user's emotional state, allowing the user to obtain useful information in making life choices.

[1407] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1408] Step 1: User request input

[1409] A user enters specific questions or concerns about their life path into a text box in a web application. For example, they might enter, "I'm thinking about changing jobs, but I'm not sure whether I should try a new job or stay at my current job." This is the input data. The device asynchronously sends the text entered by the user to the server using AJAX. This sent data is the input for the next step.

[1410] Step 2: Receiving the request

[1411] The terminal receives text input from the user and passes it to the emotion analysis means. The input data is the text information sent by the user. The server receives the text received via an AJAX request for analysis. The output is the user's input text itself.

[1412] Step 3: Analyze emotional state

[1413] The device passes the received text to the sentiment analyzer. The sentiment analyzer uses, for example, the Hugging Face transformers library and analyzes the text with the BERT model. The input is the user's text received in step 2. The BERT model tokenizes the text and infers an emotion label based on the context. The analysis results indicate that the user is feeling anxious. The analysis results are sent to the server in JSON format.

[1414] Step 4: Parsing the request

[1415] The server analyzes the sentiment analysis results received in JSON format and the request text using natural language processing technology such as SpaCy. The input is the user's text and the sentiment analysis results. The text is tokenized using the SpaCy library, and key keywords such as "job change," "challenge," and "current workplace" are extracted. The analysis results become the input data for the next step.

[1416] Step 5: Selecting a generative AI model

[1417] The server selects the most suitable generative AI model based on the request content and the results of the sentiment analysis. The inputs are the keywords and emotional state extracted in step 4. A keyword matching algorithm is used to calculate the degree of match with the tags of each celebrity model and select the most suitable celebrity model. For example, based on the keywords "challenge" and "job change," a model that provides adventurous advice is selected. The results of this selection are the input data for the next step.

[1418] Step 6: Generating Advice

[1419] The server uses the selected generative AI model to generate advice appropriate to the user's request and emotional state. The input is the selected generative AI model and a prompt. Specifically, the prompt "I'm thinking about changing jobs. I'm feeling anxious. What should I think about this?" is input into GPT-3, and the server generates the advice "Identify where your passion lies and have the courage to take on challenges without fear of risk." This is the output data.

[1420] Step 7: Verify the advice

[1421] The server uses an internal algorithm to verify the appropriateness of the generated advice. The input is the advice generated in step 6. A rule-based filter is used to check whether it contains extreme expressions or inappropriate sentences. For example, regular expressions are used to check whether specific keywords are included and whether the expression is natural in the context. Adjustments are made as necessary, and finally, advice that passes the verification is generated as output data.

[1422] Step 8: Providing advice

[1423] The server finally sends the advice that has passed verification to the terminal and displays it to the user. The input is verified advice. The terminal uses HTML and CSS to display the received advice in an easy-to-read GUI (Graphical User Interface). Important points are highlighted to make it easier for the user to understand the advice.

[1424] (Application example 2)

[1425] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1426] Currently, there are limited ways to quickly provide appropriate advice to employees in brick-and-mortar stores regarding work-related or career-related issues. It is also difficult to provide advice that takes into account the emotional state of employees, resulting in problems such as a decline in employee motivation and the accumulation of stress. Furthermore, there is no established method for ensuring the reliability of the advice provided.

[1427] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving requests from users, means for saving famous people's quotes, personal experiences, and past advice as a dataset, means for analyzing the received requests and selecting and training a generative AI model based on the necessary training dataset, means for generating advice in response to the user request using the selected generative AI model, an emotion engine for analyzing the user's emotional state, and an application for solving problems between employees and providing feedback. This makes it possible to quickly provide reliable advice that takes into account the emotional state of employees.

[1428] A "terminal" is a device that receives requests from users and provides generated advice.

[1429] "Server" means a central processing unit for storing and analyzing data, selecting and training generative AI models, and generating and providing advice.

[1430] A "celebrity" refers to a person who is widely known in society, and whose famous quotes, experiences, advice, etc. are used as a dataset.

[1431] A "dataset" is a collection of famous quotes, personal experiences, and past advice that is used as training material for generative AI models.

[1432] A "generative AI model" is an artificial intelligence model that generates advice in response to a user request based on a training dataset.

[1433] The "emotion engine" is software that analyzes the user's emotional state when they input a request and adjusts the advice content based on the results.

[1434] The "application for problem-solving and feedback among employees" is software that provides advice to employees in physical stores, taking into account their emotional state, in response to their concerns and questions.

[1435] "Internal Algorithm" means an internal computational means for validating the appropriateness of the generated advice and adjusting it if necessary.

[1436] A system for implementing the present invention includes the following elements:

[1437] Hardware and Software

[1438] Hardware: Smartphone

[1439] software:

[1440] Database: MySQL, PostgreSQL, etc.

[1441] Emotion engine: EmotionAPI

[1442] Generative AI model: OpenAI GPT-4

[1443] Application backend: Node.js

[1444] Frontend: React Native

[1445] Explanation of program processing

[1446] First, the server collects quotes, personal experiences, and past advice from famous people online and from books. This data is stored in a text database (MySQL or PostgreSQL) and cleaned under the supervision of experts. The cleaned data is organized into a training dataset, and a generative AI model (OpenAI GPT-4) is trained using this data.

[1447] Users access the application with specific concerns or questions and enter their request in text form via their smartphone. For example, they can enter something like, "I'm not sure whether I should apply for a new position or stay in my current one."

[1448] The device is equipped with an emotion engine (Emotion API) that analyzes the user's emotional state when they input a request. The emotion engine detects the user's emotions from the text input and sends the emotion analysis results to the server.

[1449] The server analyzes the received request and the results of sentiment analysis, and selects the most suitable generative AI model. This analysis is performed using syntactic analysis technology to extract keywords and important points contained in the request.

[1450] The server then uses the selected generative AI model to generate advice. The generative AI model references the learned thinking patterns and past advice of famous people to provide advice tailored to the user's concerns and emotional state. OpenAI GPT-4 is used to generate this advice.

[1451] The generated advice is then validated by an internal algorithm to ensure its appropriateness, including whether the advice is overly biased or if the wording is appropriate. If necessary, the advice is adjusted.

[1452] The final verified and adjusted advice is sent back to the terminal and provided to the user, who can then consider their own choices based on this advice.

[1453] Specific examples

[1454] Example 1: Career Choice and Sentiment Analysis

[1455] 1. The user enters the following text: "I'm not sure whether I should apply for a new position or stay in my current one."

[1456] 2. The device receives this request and uses its emotion engine to analyze that the user is feeling anxious.

[1457] 3. The device sends the request and emotional state to the server.

[1458] 4. The server selects an appropriate generative AI model based on the request and the results of sentiment analysis.

[1459] 5. The server generates advice based on the generated AI model, such as "Identify your passion and have the courage to take on challenges without fear of risk," and adds expressions to ease anxiety.

[1460] 6. The server sends the verified advice to the terminal and displays it to the user.

[1461] Prompt Sentence Examples

[1462] "I'm not sure whether I should apply for a new position or stay in my current one. How should I decide? Helplessness: I feel it. Anxiety: I feel it. Confidence: I want it."

[1463] In this way, the system provides reliable advice that takes into account the user's emotional state, enabling store staff to solve problems and provide feedback.

[1464] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1465] Step 1:

[1466] Famous quotes, personal experiences, and past advice from famous people are collected online and from books. This data is stored in text format in a database (MySQL or PostgreSQL). The collected data is then cleaned and reviewed by experts and organized into a training dataset. This ensures high-quality data is used to train generative AI models.

[1467] Input: Text data collected from online sources and books

[1468] Output: Data organized as a training dataset

[1469] Step 2:

[1470] The prepared training dataset will be used to train a generative AI model (OpenAI GPT-4), which will then learn the thought patterns and advice of celebrities, laying the foundation for generating appropriate advice in response to user requests.

[1471] Input: A curated training dataset

[1472] Output: A trained generative AI model

[1473] Step 3:

[1474] The user enters specific concerns or questions in text format via their smartphone. For example, they might enter something like, "I'm not sure whether I should apply for a new position or stay in my current one." This text data is then analyzed in the next step.

[1475] Input: The textual request entered by the user

[1476] Output: Request data in text format

[1477] Step 4:

[1478] The emotion engine (Emotion API) built into the device analyzes the user's emotional state when they input a request. The emotion engine analyzes the text data and detects the emotion the user is feeling (e.g., anxiety or helplessness). The detected emotional state is sent to the server.

[1479] Input: Request data in text format

[1480] Output: Emotional analysis results (e.g. anxiety, helplessness)

[1481] Step 5:

[1482] The server analyzes the received request and the sentiment analysis results, and performs a syntactic analysis to extract keywords and key points contained in the request. This analysis clarifies the important points of the request content.

[1483] Input: Request data in text format, sentiment analysis results

[1484] Output: Keywords and key points of the request

[1485] Step 6:

[1486] The server selects the most suitable generative AI model based on the extracted keywords and key points, and uses the selected generative AI model to generate advice in response to the user's request. The generated advice also takes into account the user's emotional state.

[1487] Input: Keywords and key points of the request, sentiment analysis results

[1488] Output: Generated advice

[1489] Step 7:

[1490] The generated advice is verified by an algorithm inside the server to check its appropriateness, checking whether the advice is excessively biased and whether appropriate wording is used, and adjusting the advice content as necessary.

[1491] Input: Generated advice

[1492] Output: Validated and adjusted advice

[1493] Step 8:

[1494] The final verified and adjusted advice is then sent back to the device and provided to the user, who can then receive the advice via their smartphone and consider their own choices.

[1495] Input: Validated and adjusted advice

[1496] Output: Advice given to the user

[1497] The above steps make it possible to provide fast, reliable advice that takes into account the emotional state of employees in physical stores in response to their concerns and questions.

[1498] 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 control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1499] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1500] 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 the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1501] The emotion identification model 59 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 an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1502] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1503] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1504] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1505] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1506] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs 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 a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1507] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1508] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1509] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1510] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1511] 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.

[1512] It is not necessary to store all 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 all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1513] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1514] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with 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). Also, the hardware resource that executes the specific processing may be a single processor.

[1515] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1516] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1517] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1518] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1519] The following is further disclosed regarding the above embodiment.

[1520] (Claim 1)

[1521] a terminal that receives a request from a user;

[1522] A server that stores famous quotes, personal experiences, and past advice as a dataset, and

[1523] A server that analyzes the received requests and selects and trains a generative AI model based on the required training dataset;

[1524] a server that generates advice in response to a user request using the selected generative AI model;

[1525] a terminal for providing the generated advice to a user;

[1526] A system including:

[1527] (Claim 2)

[1528] The dataset was formalized with the permission and supervision of the celebrities themselves.

[1529] 10. The system of claim 1, wherein the system generates reliable advice.

[1530] (Claim 3)

[1531] 10. The system of claim 1, further comprising means for verifying the appropriateness of the generated advice with an internal algorithm and adjusting it if necessary.

[1532] "Example 1"

[1533] (Claim 1)

[1534] an information processing device that receives a request from a user;

[1535] a data storage device that stores quotes and experiences using a database;

[1536] an analysis device that analyzes the received request using a natural language processing tool;

[1537] A learning device that selects and trains the optimal generative AI model based on the analysis results;

[1538] a generating device that generates advice in response to a user request using the selected generative AI model;

[1539] a display device that provides the generated advice to the user;

[1540] A system including:

[1541] (Claim 2)

[1542] The system of claim 1, which organizes and curates a dataset to generate reliable advice.

[1543] (Claim 3)

[1544] 2. The system according to claim 1, further comprising a verification device that verifies the appropriateness of the generated advice using an internal algorithm.

[1545] "Application Example 1"

[1546] (Claim 1)

[1547] a terminal that receives a request from a user;

[1548] An information processing device that stores famous people's words, experiences, and past advice as a data set;

[1549] an information processing device that analyzes the received request, selects and trains a generative AI model based on the required training data set;

[1550] an information processing device that generates advice in response to a user request using the selected generative AI model;

[1551] means for enabling customers to obtain product advice using a terminal in a physical store;

[1552] a terminal for providing the generated advice to a user;

[1553] A system including:

[1554] (Claim 2)

[1555] The system of claim 1, wherein the dataset is formulated with the permission and supervision of the celebrity himself / herself to generate reliable advice.

[1556] (Claim 3)

[1557] 10. The system of claim 1, further comprising means for verifying the appropriateness of the generated advice with an internal algorithm and adjusting it if necessary.

[1558] "Example 2: Combining Emotion Engines"

[1559] (Claim 1)

[1560] a terminal that receives a request from a user;

[1561] A server that stores famous quotes, experiences, and past advice as a dataset,

[1562] A means for analyzing received requests and extracting necessary keywords and key points using natural language processing technology;

[1563] emotion analysis means for recognizing the user's emotional state and adjusting the content of advice based on the analysis result;

[1564] A server that selects and trains a generative AI model based on the required training dataset;

[1565] a server that generates advice in response to a user request using the selected generative AI model;

[1566] A means to verify the appropriateness of the advice generated using internal algorithms and adjust it as necessary;

[1567] a terminal for providing the generated advice to a user;

[1568] A system including:

[1569] (Claim 2)

[1570] The system of claim 1, wherein the dataset is formulated with the permission and supervision of the celebrity himself / herself and generates reliable advice.

[1571] (Claim 3)

[1572] 10. The system of claim 1, further comprising means for verifying the appropriateness of the generated advice with an internal algorithm and adjusting it if necessary.

[1573] "Application example 2 when combining emotion engines"

[1574] (Claim 1)

[1575] a terminal that receives a request from a user;

[1576] A server that stores famous quotes, personal experiences, and past advice as a dataset, and

[1577] A server that analyzes the received requests and selects and trains a generative AI model based on the required training dataset;

[1578] a server that generates advice in response to a user request using the selected generative AI model;

[1579] a terminal for providing the generated advice to a user;

[1580] an emotion engine that analyzes the user's emotional state;

[1581] Applications aimed at problem-solving and feedback among employees,

[1582] A system including:

[1583] (Claim 2)

[1584] The dataset was formalized with the permission and supervision of the celebrities themselves.

[1585] 10. The system of claim 1, wherein the system generates reliable advice.

[1586] (Claim 3)

[1587] 10. The system of claim 1, further comprising means for verifying the appropriateness of the generated advice with an internal algorithm and adjusting it if necessary. [Explanation of symbols]

[1588] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a terminal that receives a request from a user; A server that stores famous quotes, personal experiences, and past advice as a dataset, and A server that analyzes the received requests and selects and trains a generative AI model based on the required training dataset; a server that generates advice in response to a user request using the selected generative AI model; a terminal for providing the generated advice to a user; A system including:

2. The dataset was formalized with the permission and supervision of the celebrities themselves. The system of claim 1 , wherein the system generates reliable advice.

3. 10. The system of claim 1, further comprising means for verifying the appropriateness of the generated advice with an internal algorithm and adjusting it if necessary.

Citation Information

Patent Citations

  • Persona chatbot control method and system

    JP2022180282A