System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving an input of a user; means for analyzing the input and passing the input to a generative artificial intelligence model; means for generating an answer to the input of the user using the generative artificial intelligence model; means for returning the generated answer to a user terminal; and means for receiving feedback from the user and using the feedback for learning of the generative artificial intelligence model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In today's educational environment, providing appropriate and immediate learning support to individual students is a challenge, especially in cram schools and online classes. This challenge stems from the current situation where teachers' workloads are increasing, making it difficult to provide sufficient support to each student. Furthermore, online classes make it difficult to hold real-time Q&A sessions and grasp students' learning progress. A new system is needed to solve these issues and provide effective and efficient learning support. [Means for solving the problem]

[0005] This invention provides a system that includes a means for accepting input from a user (student or teacher), a means for analyzing the input and passing it to a generative AI model, a means for generating an answer to the user's input using the generative AI model, a means for returning the generated answer to the user's terminal, and a means for receiving feedback from the user and using it for training the generative AI model. This system also includes a means for storing the feedback in a database and a means for the generative AI model to use natural language processing technology. This system provides quick and appropriate learning support to individual students, reduces the burden on teachers, and enables effective learning support even in online classes.

[0006] "User" refers to an individual, such as a student or teacher, who uses this system to provide learning support or answer questions.

[0007] "Input interface" refers to a software platform, such as a web application or mobile application, through which a user inputs a question or request.

[0008] "Server" refers to the central processing unit that receives input from users and passes it to the generative AI model for processing.

[0009] A "generative artificial intelligence model" refers to an AI algorithm that uses natural language processing techniques to generate answers to user input.

[0010] "Answer" refers to text data generated by a generative artificial intelligence model that contains information and explanations in response to a user's question.

[0011] "User terminal" refers to a device such as a computer or smartphone that a user uses to access the system, enter questions, and check answers.

[0012] "Feedback" refers to a user providing a rating or opinion on a generated answer.

[0013] "Database" means the data storage system for storing and managing Feedback and other related data.

[0014] "Natural language processing technology" refers to technology that enables AI to understand human language, analyze its meaning, and generate appropriate responses. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention is a learning support system in which a user (a student or a teacher) inputs a question and an answer to the question is provided using a generative AI model. Specific embodiments of the present invention are described below.

[0037] 1. User inputs a question

[0038] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0039] 2. Submit your question

[0040] The device has a function to send questions entered by the user to the server, and the entered data is sent to the server via a specific API endpoint.

[0041] 3. Receiving and analyzing questions

[0042] The server receives questions sent from the user's device, analyzes the data, and passes it to the generative AI model. Specifically, it converts the input text data into a format that the AI ​​model can understand.

[0043] 4. Answer Generation

[0044] The server then passes the analyzed question data to a generative AI model, which processes it to generate the optimal answer to the question. This generative AI model uses natural language processing technology and is capable of generating advanced answers based on past data and learning results.

[0045] 5. Returning the Response

[0046] The server then sends the generated answer back to the user's device, which may include, for example, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0047] 6. View Answers

[0048] The terminal displays the answer returned from the server to the user, allowing the user to confirm an immediate and appropriate answer to the question.

[0049] 7. Providing Feedback

[0050] Users can provide ratings and opinions on generated answers based on such things as whether the answer was helpful or easy to understand.

[0051] 8. Sending and Saving Feedback

[0052] The device sends user feedback to the server, which stores it in a database and uses it to train and improve the generative AI model, allowing the system to continuously improve its accuracy and quality.

[0053] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." The user can review this answer and provide a high rating if they find it specific and easy to understand. This feedback is stored in the system and used to generate future answers.

[0054] In this way, the present invention is a system that allows users to obtain immediate and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning.

[0055] The processing flow will be explained below.

[0056] Program processing flow

[0057] The specific processing flow of this system will be explained below step by step.

[0058] Step 1:

[0059] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0060] Step 2:

[0061] The device sends the data of the question entered by the user to the server using a dedicated API endpoint, and the entered question is sent in text data format.

[0062] Step 3:

[0063] The server receives questions sent from the user's device. The received question data is preprocessed for analysis and prepared for passing to the generative AI model. Specifically, the text data is tokenized and converted into a format that is easy for the AI ​​model to understand.

[0064] Step 4:

[0065] The server inputs the preprocessed question data into a generative AI model, which generates the optimal answer. This generative AI model uses natural language processing technology to generate answers based on past data and learning results.

[0066] Step 5:

[0067] The server then sends the generated answer back to the user's device, such as, "Photosynthesis is a process primarily performed by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0068] Step 6:

[0069] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers and deepen their understanding of the learning content.

[0070] Step 7:

[0071] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[0072] Step 8:

[0073] The terminal transmits the feedback from the user to the server, and the feedback data is transmitted in the form of ratings and comments.

[0074] Step 9:

[0075] The server stores the received feedback in a database. This feedback data is used to train and improve the generative AI model. Accumulating feedback data improves the accuracy and quality of the entire system.

[0076] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, maximizing the effectiveness of learning support. It also reduces the burden on teachers and provides learning support tailored to each student.

[0077] Example 1

[0078] 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."

[0079] In modern learning support systems, it is not easy to provide immediate and appropriate answers to questions submitted by users. Furthermore, improving the quality of generated answers and learning models by utilizing subsequent feedback are also important issues. Therefore, a system that can do this efficiently is needed.

[0080] 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.

[0081] In this invention, the server includes means for inputting a question through a user's terminal, means for transmitting the question to the server, means for receiving and analyzing the question at the server and passing it as a prompt to the generative AI model, means for generating an answer to the question using the generative AI model, means for returning the generated answer to the user's terminal, means for displaying the generated answer on the user's terminal, means for receiving feedback from the user, means for transmitting and storing the feedback to the server, and means for using the feedback to train the generative AI model. This allows users to receive instant, high-quality answers to their questions, and the system is continuously improved based on the feedback.

[0082] A "user's terminal" is an electronic device, such as a PC or smartphone, that a user uses to input information.

[0083] A "question" is a textual inquiry that a user enters into the system.

[0084] The "server" is a central processing unit that receives and analyzes instructions sent from the user's device and passes them to the generative AI model.

[0085] A "generative AI model" refers to an artificial intelligence algorithm that uses natural language processing technology to generate optimal answers to input questions.

[0086] A "prompt" is a pre-processed textual instruction that is passed to a generative AI model.

[0087] An "answer" is a text response generated by a generative AI model in response to a user's question.

[0088] "Feedback" refers to the ratings and opinions that users provide on generated answers.

[0089] A "database" is a data storage system for storing and managing data such as feedback.

[0090] "Natural language processing technology" is a computational technology for analyzing, understanding, and generating language that humans use on a daily basis.

[0091] This invention is a learning support system that uses a generative AI model to provide appropriate answers to questions entered by a user. This system consists of multiple components, including a user terminal, a server, and a generative AI model.

[0092] 1. User inputs a question

[0093] Users use a device such as a PC or smartphone to enter questions into a dedicated input interface. Specifically, they open a web browser, access the system's web page, and enter a question into the input form, such as "Please tell me about the process of photosynthesis."

[0094] 2. Submit your question

[0095] The device sends the question entered by the user to the server as an HTTP POST request via an API endpoint. The data is transferred in JSON format, for example:

[0096] json

[0097] {

[0098] "question": "What is the process of photosynthesis?"

[0099] }

[0100] 3. Receiving and analyzing questions

[0101] The server receives the HTTP POST request sent from the device and analyzes the question data contained therein. This analysis includes normalizing the text data and removing unnecessary whitespace and special characters. This converts the text data into a format that the generative AI model can understand.

[0102] 4. Answer Generation

[0103] The server then passes the analyzed question data to a generative AI model as a prompt to generate the optimal answer. This generative AI model uses natural language processing technology, such as OpenAI's GPT-3, which runs on the cloud. The prompt is entered in the following format:

[0104] text

[0105] "Tell me about the process of photosynthesis."

[0106] Generative AI models take into account past data and context to generate answers like:

[0107] text

[0108] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0109] 5. Returning and Displaying Answers

[0110] The server returns the generated answer in JSON format to the user's device as an HTTP response. The device displays the answer received from the server on the screen and tells the user, for example, as follows:

[0111] text

[0112] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0113] 6. Providing and Submitting Feedback

[0114] The user inputs their rating and opinion for the generated answer. For example, they select a rating such as "very helpful" in the rating form. The device sends this rating to the server as an HTTP POST request. An example of the data to be sent is as follows:

[0115] json

[0116] {

[0117] "question": "Explain the process of photosynthesis",

[0118] "answer": "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen.",

[0119] "feedback": "Very helpful"

[0120] }

[0121] 7. Storage and Use of Feedback

[0122] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance and answer accuracy.

[0123] In this way, the system of the present invention allows users to receive instant, high-quality answers to their questions and is continuously improved based on feedback, thereby maximizing the effectiveness of learning support, reducing the burden on teachers, and ultimately improving the quality of learning.

[0124] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0125] Step 1:

[0126] A user inputs a question into a dedicated input interface using a device such as a PC or smartphone. For example, the input question might be text such as "Please tell me about the process of photosynthesis." This input is completed when the user enters the text into an HTML form element and presses the submit button.

[0127] Input: User-provided question text

[0128] Output: Question text in terminal input form

[0129] Step 2:

[0130] The device sends the question entered by the user to the server by sending an HTTP POST request to the API endpoint. The question text is encoded in JSON format and sent.

[0131] Input: User question text

[0132] Output: JSON formatted data sent to the server

[0133] Step 3:

[0134] The server receives an HTTP POST request sent from the device, which contains the user's question. At this stage, the server extracts the question text from the request and normalizes it, removing unnecessary whitespace and special characters.

[0135] Input: JSON format data received from the terminal

[0136] Output: Normalized question text

[0137] Step 4:

[0138] The server passes the normalized question text as a prompt to a generative AI model, which uses natural language processing techniques, for example. The server sends the prompt to the generative AI model, which then generates an answer.

[0139] Input: Normalized question text

[0140] Output: Generated answer text

[0141] Step 5:

[0142] The server packages the answer received from the generative AI model in JSON format and returns it to the user's device as an HTTP response, which includes the generated answer text.

[0143] Input: Answer text from the generative AI model

[0144] Output: JSON formatted response data

[0145] Step 6:

[0146] The terminal processes the HTTP response received from the server and displays the generated answer to the user in HTML format. Using JavaScript, the answer text is inserted into the specified DOM element on the browser.

[0147] Input: JSON format response data from the server

[0148] Output: Answer text displayed in the browser

[0149] Step 7:

[0150] The user can enter feedback on the displayed answer by using the evaluation form to select a rating such as "helpful" or "easy to understand" and enter comments if necessary.

[0151] Input: User feedback on the generated answer

[0152] Output: Feedback text

[0153] Step 8:

[0154] The device sends the user-entered feedback to the server, again using an HTTP POST request, with the feedback encoded in JSON format.

[0155] Input: User feedback text

[0156] Output: Feedback data sent to the server in JSON format.

[0157] Step 9:

[0158] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance.

[0159] Input: Feedback data received from the device in JSON format

[0160] Output: Feedback data stored in a database

[0161] (Application example 1)

[0162] 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."

[0163] In traditional brick-and-mortar stores, when customers wanted detailed information about a product or a guided tour of the store, it was difficult to provide them with the information quickly and accurately. Furthermore, because the store required the assistance of an employee, responses were often delayed during busy periods, resulting in lower customer satisfaction. Furthermore, the quality and speed of responses to customer questions depended on the experience and knowledge of the employee, which meant that consistent service could be inconsistent. To solve these issues, a system was needed that would allow customers to input questions and receive instant answers.

[0164] 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.

[0165] In this invention, the server includes a means for accepting user input, a means for analyzing the input and passing it to the generative AI model, and a means for returning the generated answer to the user terminal, thereby enabling quick and accurate answers to questions about product explanations, store guides, and recipe suggestions in a physical store to be provided.

[0166] The "means for accepting user input" is a function that allows a user to input questions or requests to the terminal via a dedicated input interface.

[0167] The "means for analyzing input and passing it to the generative artificial intelligence model" is a processing function for analyzing data input by a user, converting it into a format that can be understood by the artificial intelligence model, and passing it on.

[0168] "Means for generating answers to user input using a generative artificial intelligence model" refers to algorithms and techniques that allow the artificial intelligence model to generate optimal answers based on analyzed input data.

[0169] "Means for returning the generated answer to the user terminal" refers to communication and display technology for transferring and displaying the answer generated by the artificial intelligence model to the user's terminal.

[0170] "Means for receiving feedback from users and using it to train the generative AI model" refers to data collection and learning technology that collects evaluations and opinions on answers provided by users and uses them to improve and enhance the accuracy of the AI ​​model.

[0171] "Means of providing answers to questions regarding product descriptions, store guidance, and recipe suggestions within a physical store" refers to functions and technologies that allow customers to ask questions regarding product and store information and recipe suggestions in a physical store, and for an AI model to provide immediate and appropriate answers.

[0172] "Means for storing data in a database" refers to the storage systems and technologies used to organize and securely store collected data and feedback.

[0173] "Using natural language processing technology" refers to the process of analyzing text entered by a user and generating an answer using algorithms and technologies for understanding and generating human language.

[0174] A specific embodiment of the present invention will be described in detail below. This invention is a system that allows users (customers) to ask questions about products, store information, and recipe suggestions in a physical store and receive automatic answers.

[0175] System Configuration

[0176] Hardware

[0177] The server uses a virtual machine on the cloud (e.g., AWS EC2, Google Cloud VM), and the user's device is a smartphone (iOS / Android) or PC.

[0178] software

[0179] The programming language used is Python.

[0180] Use Flask as a web framework.

[0181] To analyze questions and generate answers, we use generative AI models provided by OpenAI (e.g., GPT-3).

[0182] Detailed System Description

[0183] User Input

[0184] Users input questions using a dedicated input interface on their smartphone or PC, such as "Where is this wine produced?" or "Please tell me some recipes using this cheese."

[0185] Submit a Question

[0186] The device sends the entered question to a server in the cloud via a specific API endpoint.

[0187] Receiving and parsing questions

[0188] The server receives questions sent by users, analyzes the data, and passes it to a generative AI model (e.g., GPT-3). Specifically, it converts the text data into a format that the AI ​​model can understand.

[0189] Generate answers

[0190] The server uses the analyzed question data to input the generative AI model and generate the optimal answer, such as "This wine is produced in the Bordeaux region of France."

[0191] Returning the answer

[0192] The server returns the generated answer to the user terminal, which displays the returned answer on the screen for the user to confirm.

[0193] Providing and storing feedback

[0194] Users can provide ratings and opinions on generated answers based on their helpfulness and ease of understanding. Feedback is sent to the server and stored in a database. This information is used to continuously learn and improve the generative AI model.

[0195] Specific examples

[0196] For example, if a user types "Where is this wine from?" the system generates the following prompt:

[0197] Prompt Sentence Examples

[0198] Prompt: "Q: Where is this wine from?\nA:"

[0199] The server passes this prompt to the generative AI model, which then generates the answer, "This wine is produced in the Bordeaux region of France." The user receives this information through their device and provides a high rating if they are satisfied with the answer.

[0200] The above is a detailed description of an embodiment of the present invention. The present invention is a system that can improve convenience for customers in a physical store and reduce the burden on employees.

[0201] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0202] Step 1:

[0203] The user inputs a question using the input interface on the terminal. For example, the user can input the question, "Where is this wine produced?" The input data is in text format, and the terminal sends it to the server.

[0204] Step 2:

[0205] The device sends the entered question to a server on the cloud via a specific API endpoint. The sent data is in text format. The server receives this data.

[0206] Step 3:

[0207] The server receives the question sent by the user and analyzes the text data. The analyzed data is converted into a format that can be understood by a generative AI model (e.g., GPT-3). In this step, a prompt sentence is generated. For example, the format is "Prompt: "Q: Where is this wine produced?\nA:"".

[0208] Step 4:

[0209] The server inputs the generated prompt sentence into the generative AI model, which then generates the optimal answer based on the prompt. Data processing involves converting the input text into vector format and applying natural language processing algorithms to generate the answer.

[0210] Step 5:

[0211] The server receives the answer generated by the generative AI model and prepares it to be sent back. The generated answer is in text format. For example, the answer generated might be, "This wine is produced in the Bordeaux region of France."

[0212] Step 6:

[0213] The server sends the generated answer back to the user's terminal, which displays the received answer on the screen, allowing the user to check the answer to the question.

[0214] Step 7:

[0215] Users provide ratings and opinions on the generated answers, such as whether the answer was helpful or easy to understand. The rating data is entered in text format or multiple-choice format.

[0216] Step 8:

[0217] The device sends user feedback to a server, which receives it and stores it in a database. The stored evaluation data is used to train and improve the generative AI model in the future.

[0218] The above is the specific processing flow of the program that realizes this application example. This flow makes it possible to provide quick and accurate answers to customer questions in a physical store.

[0219] 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.

[0220] The present invention is a learning support system in which users (students or teachers) input questions and answers are provided using a generative AI model. Furthermore, the present invention achieves more accurate learning support by combining an emotion engine that recognizes the user's emotions. Specific embodiments of the present invention are described below.

[0221] 1. User inputs a question

[0222] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0223] 2. Starting Emotion Recognition

[0224] The device recognizes the user's emotions by using facial expressions and voice data when the user inputs a question. This emotion recognition is performed using an emotion engine.

[0225] 3. Sending questions and emotion data

[0226] The device sends the question entered by the user and the accompanying emotion data to the server. The entered data is sent as question text data and emotion data from the emotion engine.

[0227] 4. Receiving and analyzing questions

[0228] The server receives and analyzes questions and emotion data sent from the user's device. The question data is preprocessed and passed to the generative AI model, and the emotion data is used to generate and adjust the answer.

[0229] 5. Generate and refine answers

[0230] The server uses a generative AI model to generate optimal answers to questions. Furthermore, it adjusts the tone and content of the answer based on the user's emotions, based on emotional data from the emotion engine. For example, if the user looks anxious, a more friendly and polite answer will be generated.

[0231] 6. Returning the Response

[0232] The server then sends a tailored response back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Easy to understand, right?"

[0233] 7. View Answers

[0234] The terminal displays the answers returned from the server to the user, who can then check the adjusted answers and deepen their understanding of the learning content.

[0235] 8. Providing Feedback

[0236] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[0237] 9. Sending and Saving Feedback

[0238] The device sends user feedback to the server, which stores it in a database and uses it to subsequently train and improve the generative AI model and emotion engine, thereby improving the accuracy and quality of the entire system.

[0239] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with, "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." Furthermore, if the user looks anxious, the system will include additional explanation such as, "It's okay, let's understand it well." The user will confirm this response and appreciate the detailed and easy-to-understand explanation. In this way, by utilizing the emotion engine, it is possible to provide customized learning support according to the user's emotions.

[0240] In this way, the present invention is a system that allows users to receive prompt and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning. Furthermore, by using an emotion engine, support that takes into consideration the user's emotions can be realized, providing even greater satisfaction and understanding.

[0241] The processing flow will be explained below.

[0242] Program processing flow

[0243] Step 1:

[0244] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0245] Step 2:

[0246] The device receives the question entered by the user as text data and simultaneously acquires emotion data from the user's facial expressions and voice. The emotion data is used to identify the user's emotion using video and audio analysis approaches.

[0247] Step 3:

[0248] The device transmits the acquired question data and emotion data to the server using a secure communication protocol.

[0249] Step 4:

[0250] The server receives question data and emotion data sent from the user's device. After receiving the data, it performs preprocessing such as tokenizing the text data, and prepares it for passing to the generative AI model and emotion engine.

[0251] Step 5:

[0252] The server uses a generative artificial intelligence model to generate optimal answers based on the question data. The generated answers are based on the underlying learning content.

[0253] Step 6:

[0254] The server uses an emotion engine to tailor the generated responses, specifically adjusting the tone and expression of the responses based on the user's emotion data. For example, if the user expresses anxiety, the responses will be more kind and polite.

[0255] Step 7:

[0256] The server then sends tailored answers back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Was this easy to understand?"

[0257] Step 8:

[0258] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers to deepen their understanding of the learning content.

[0259] Step 9:

[0260] Users provide ratings and opinions (feedback) on the generated answers, such as whether the answers were helpful, easy to understand, or whether they made them feel a certain emotion.

[0261] Step 10:

[0262] The device sends user feedback and emotion data to the server, which will be used to improve the system in the future.

[0263] Step 11:

[0264] The server stores the received feedback and emotion data in a database. This data is used to train and improve the generative AI model and emotion engine. Accumulating data improves the accuracy and quality of the system.

[0265] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, and by using an emotion engine, it provides personalized learning support according to the user's emotions. Furthermore, feedback data can be used to continuously improve the system.

[0266] Example 2

[0267] 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."

[0268] Conventional learning support systems can use artificial intelligence (AI) to provide answers to user questions, but do not generate answers that take the user's emotions into account. This makes it difficult to address the anxiety and confusion users feel while learning, which can reduce the quality and effectiveness of learning. Furthermore, there are also insufficient mechanisms for appropriately accumulating user feedback and reflecting it in subsequent system improvements. The present invention aims to solve these problems.

[0269] 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 means for accepting user input, means for recognizing the user's emotion, means for analyzing the input and emotion data and passing it to a generative AI model, means for generating a response to the user's input using the generative AI model and adjusting the response based on the user's emotion, means for returning the generated response to the user terminal, and means for receiving feedback from the user and using it for training the generative AI model and the emotion recognition means. As a result, by recognizing and responding to the user's emotion, it is possible to provide more appropriate and effective learning support and improve the accuracy and quality of the entire system based on the feedback.

[0270] The "means for accepting user input" refers to a means by which a user inputs questions or information related to learning in the form of text, voice, or the like.

[0271] The "means for recognizing the user's emotions" is a means for analyzing the user's facial expressions and voice and identifying the emotions the user is feeling.

[0272] The "means for analyzing the input and emotion data and passing it to the generative AI model" refers to a means for analyzing the user's question text and voice, as well as the recognized emotion data, and passing it to the generative AI model.

[0273] "Means for using a generative artificial intelligence model to generate a response to a user's input and adjusting the response based on the user's emotions" refers to means for using a generative AI model to generate an optimal response to a user's question and further adjusting the content and tone of the response depending on the user's emotions.

[0274] The "means for returning the generated answer to the user terminal" refers to a means for transmitting the generated and adjusted answer to the user's terminal.

[0275] "Means for receiving feedback from users and using it to train the generative artificial intelligence model and emotion recognition means" refers to means for receiving evaluations and opinions on answers provided by users and using them to improve the performance of the generative artificial intelligence model and emotion recognition engine.

[0276] "Means for storing in a database" means a means for permanently recording user feedback and other relevant data for later analysis and system improvement.

[0277] "Using natural language processing technology" means using technology that applies natural language processing algorithms to the analysis and generation of text data.

[0278] MODE FOR CARRYING OUT THE INVENTION

[0279] The present invention is a system that uses a generative AI model to provide answers to questions about learning entered by a user, and further combines it with an emotion engine to enable customized learning support according to emotions. The following describes in detail the embodiments for implementing this system.

[0280] 1. System Configuration

[0281] This system is broadly composed of a user's device, a server, and a cloud service that includes an emotion engine and a generative AI model.

[0282] User's device

[0283] Users access a dedicated learning support input interface using devices such as PCs or smartphones. This interface is web browser-based software (e.g., implemented using React.js) and is intuitive to operate.

[0284] server

[0285] The server is the central part that receives input from the user and performs the necessary analysis. It is powered by Node.js and Express.js and consists of the following components:

[0286] Input receiving component: Receives question text and sentiment data from the user.

[0287] Preprocessing component: Performs preprocessing such as tokenizing the question text and removing unnecessary symbols.

[0288] Generative AI model invocation component: Invokes a generative AI model, such as OpenAI's GPT-4, to generate answers to questions.

[0289] Emotion adjustment component: Adjusts the generated answers based on the user's emotional data.

[0290] Feedback collection component: receives feedback from users and stores it in a database.

[0291] Emotion Engine

[0292] The emotion engine analyzes the user's facial expressions and voice to identify the emotions they are feeling, using the Emotion API from Microsoft Azure Cognitive Services and the Cloud Speech-to-Text API from Google Cloud.

[0293] Description of software and hardware used

[0294] Webcam and microphone: Hardware that captures a user's facial expressions and records their voice.

[0295] Microsoft Azure Cognitive Services Emotion API: Software for analyzing facial expression data and recognizing emotions.

[0296] Google Cloud Speech-to-Text API: Software for analyzing voice data and recognizing emotions.

[0297] OpenAI GPT-4: A generative AI model that generates answers to questions.

[0298] Node.js and Express.js: Software for performing server-side processing.

[0299] Example of a system

[0300] As a concrete example, if a user types the question "Tell me about the process of photosynthesis," the system will do the following:

[0301] 1. The user types "Please tell me about the process of photosynthesis" into the PC's input interface.

[0302] 2. The device uses a webcam and microphone to capture the user's facial expressions and voice and sends the data to the emotion engine.

[0303] 3. The emotion engine analyzes the captured data and identifies the user's emotion as "anxiety."

[0304] 4. The device sends the entered question and emotion data to the server.

[0305] 5. The server preprocesses the received data and passes it to a generative AI model to generate an answer.

[0306] 6. After the answer is generated, the server takes into account the sentiment data and adds additional encouragement to the answer, such as "It's okay, you'll understand."

[0307] 7. The server sends the tailored response back to the user's terminal.

[0308] 8. The terminal displays the answer for the user to confirm.

[0309] 9. The user rates the answer and provides feedback on whether it was helpful.

[0310] 10. The device sends the feedback to the server, which stores it in a database.

[0311] Prompt Sentence Examples

[0312] Examples of prompts include:

[0313] "Please explain the process of photosynthesis in detail. The user looks anxious."

[0314] In this way, the present invention can provide customized learning assistance that takes into account the user's emotions, improving the quality of learning.

[0315] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0316] Step 1: User enters question

[0317] The user inputs a question related to the study into the input interface of the terminal. The input question text is sent to the interface. For example, by inputting "Please tell me about the process of photosynthesis," the text data is sent to the terminal.

[0318] Step 2: Start Recognizing Emotions

[0319] The device captures the user's facial expressions with a webcam and records their voice with a microphone. The captured data (image data and voice data) is sent to an emotion engine. Specific emotion engines used here include the Emotion API and Cloud Speech-to-Text API. The emotion engine then analyzes the input data and generates emotion data.

[0320] Step 3: Submit your question and sentiment data

[0321] The device sends the entered question text and analyzed emotion data in JSON format to the server. The sent data includes, for example, the following format:

[0322] json

[0323] {

[0324] "question": "Please explain the process of photosynthesis.",

[0325] "emotion": {

[0326] "type": "anxious",

[0327] "confidence": 0.85

[0328] }

[0329] }

[0330] This causes the data to be sent to the server for further processing.

[0331] Step 4: Receive and analyze question and sentiment data

[0332] The server receives the data sent from the device. The received data is divided into question text and emotion data. The server tokenizes the question text and performs preprocessing to remove unnecessary symbols, generating data for analysis. The tokenized question text is then sent to the subsequent generative AI model.

[0333] Step 5: Generate and refine answers

[0334] The server uses a generative AI model to generate answers to questions. Specifically, it invokes OpenAI's GPT-4 model and retrieves the generated answer. For example, it might generate an answer like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen." It then adjusts the tone and content of the answer based on emotional data. If the user shows signs of anxiety, it might add a supporting sentence like, "It's okay, let's understand it."

[0335] Step 6: Submit your response

[0336] The server encodes the adjusted answer in JSON format and returns it to the user's device as an HTTP response, which sends the adjusted answer to the user's device.

[0337] Step 7: View your answers

[0338] The device decodes the response data received from the server and displays it to the user. A React.js component is used for display, allowing the user to view the adjusted response in their browser. For example, the device might display something like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen. It's okay, just make sure you understand it."

[0339] Step 8: Provide feedback

[0340] Users can provide feedback on the displayed answers through a special evaluation form, which includes options such as "helpful" or "confusing."

[0341] Step 9: Submit and save your feedback

[0342] The device encodes the user-provided feedback in JSON format and sends it to the server, which then receives the feedback and stores it in a database. This allows the feedback data to be accumulated and used for future training of the generative AI model and emotion engine.

[0343] (Application example 2)

[0344] 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."

[0345] Conventional learning support systems and customer service systems only provide uniform answers to questions entered by users (students or store clerks), and have the problem of not responding appropriately to the user's emotional state. This can result in insufficient learning benefits or customer service experience, and can lead to a decrease in user satisfaction. Therefore, there is a need for a system that can recognize the user's emotional state in real time and provide appropriate answers based on that.

[0346] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting user input, means for analyzing the input and passing it to a generative AI model, means for generating a response to the user input using the generative AI model, means for returning the generated response to the user terminal, means for receiving feedback from the user and using it for training the generative AI model, means for acquiring user emotion data using emotion recognition means built into the smart device, and means for analyzing the emotion data and adjusting the tone and content of the response. This makes it possible to generate customized responses according to the user's emotional state, thereby improving the quality of learning support and customer service support.

[0347] "User" refers to anyone who uses the system, and specifically includes learners and store clerks who serve customers.

[0348] "Means for accepting input" refers to the interface or device through which a user enters questions or instructions.

[0349] "Means for analyzing and passing to the generative artificial intelligence model" refers to the process for analyzing input data, converting it into an appropriate format, and passing it to the generative artificial intelligence model.

[0350] A "generative artificial intelligence model" refers to an AI model that uses natural language processing technology to generate answers to user input.

[0351] "Means for generating an answer" refers to a function that uses a generative artificial intelligence model to generate the optimal answer to a user's question.

[0352] "Means for returning the generated answer to the user device" refers to the function of sending the answer generated by AI to the device used by the user.

[0353] "Means of receiving feedback and using it for learning" refers to the function of collecting user evaluations and opinions and using them to improve the performance of the generative AI model.

[0354] A "smart device" is a device equipped with communication and computing capabilities, and specifically includes smart glasses, smartphones, and head-mounted displays.

[0355] "Emotion recognition means" refers to technology that uses a camera or microphone installed on the device to recognize emotions from the user's facial expressions and voice.

[0356] "Emotion data" refers to data that indicates the user's emotional state, and includes, for example, emotional states such as joy, anxiety, and anger.

[0357] "Means for adjusting the tone and content of responses" refers to a function that appropriately adjusts the expression and content of generated responses based on emotional data.

[0358] The present invention is embodied as a customer service support system for brick-and-mortar stores. Through an application using smart glasses called "Smart Tutor for Retail," the system provides quick and accurate answers to customer questions. An embodiment of this system will be described in detail below.

[0359] 1. User inputs a question

[0360] The user (store clerk) wears smart glasses and receives questions from customers as voice. For example, if a customer asks, "Please tell me more about this product," the smart glasses' high-performance microphone captures the voice.

[0361] 2. Voice Recognition

[0362] The captured voice data is converted into text by the voice recognition software in the smart glasses, using the Google Cloud Speech-to-Text API.

[0363] 3. Starting Emotion Recognition

[0364] The smart glasses use a built-in camera and microphone to capture the user's facial expressions and voice, thereby collecting emotional data. Emotion recognition is performed using Microsoft Azure Face API and voice analysis software.

[0365] 4. Sending questions and emotion data

[0366] The smart glasses send the converted text data and emotion data to a cloud server, where they are stored via GCP (Google Cloud Platform).

[0367] 5. Parsing Questions and Generating Answers

[0368] The server analyzes the question text using natural language processing (NLP) techniques to generate appropriate answers. The generative AI model used here is OpenAI's GPT-4. The generated answers are adjusted in tone and content based on emotional data.

[0369] 6. Returning and Displaying Tailored Responses

[0370] The server sends the adjusted answer back to the smart glasses, which display the answer on a display for the store clerk to see and also provide an audible response.

[0371] Examples and prompts

[0372] As a concrete example, if a customer asks, "Please tell me more about this product," the process will be as follows:

[0373] Example 1:

[0374] Customer: "Please give me a detailed description of this product."

[0375] Store clerk (through smart glasses): [Camera and microphone capture customer's question]

[0376] System: "This product is made using the latest technology and is particularly durable. For detailed characteristics, please see this URL."

[0377] Also, if the store clerk looks anxious, a more helpful and detailed explanation will be added.

[0378] Example prompt sentence:

[0379] User Question: "Can you give me a detailed description of this product?"

[0380] Prompt to generative AI model: "A customer has asked for a detailed description of a product. Please provide detailed information about the product's features, benefits, and usage, but keep it concise so that the sales associate can easily understand."

[0381] The key feature of this invention is that it recognizes the user's emotions in real time and provides customized responses based on those emotions. This improves the quality of customer service support and increases user satisfaction. In addition, the generative AI model is trained based on the collected feedback, improving the performance of the entire system.

[0382] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0383] Step 1:

[0384] (User inputs question)

[0385] The user (store clerk) wears smart glasses and receives voice questions from customers. For example, the input is voice data such as "Please tell me more about this product."

[0386] Step 2:

[0387] (Voice Recognition)

[0388] The microphone built into the smart glasses captures the received voice data, and then uses voice recognition software (Google Cloud Speech-to-Text API) to convert the voice data into text data. The input is voice data, and the output is text data.

[0389] Step 3:

[0390] (Start of emotion recognition)

[0391] The smart glasses use a built-in camera and microphone to capture the user's facial and voice data, and use the Microsoft Azure Face API to collect the user's emotion data from this data. The input is the captured facial and voice data, and the output is emotion data.

[0392] Step 4:

[0393] (Submitting questions and emotion data)

[0394] The smart glasses send text data and emotion data to a cloud server (GCP). The input is text data and emotion data, and the output is that this data is sent to the cloud server.

[0395] Step 5:

[0396] (Question analysis and answer generation)

[0397] The server analyzes the text data using natural language processing (NLP) techniques. The analyzed question is passed to a generative AI model (e.g., OpenAI's GPT-4) to generate the optimal answer. The tone and content of the generated answer are adjusted based on the emotional data. The input is the text data and emotional data, and the output is the generated answer.

[0398] Step 6:

[0399] (Return and display of adjusted answers)

[0400] The server sends the adjusted answer back to the smart glasses, which display the answer on their display and also speak it aloud. The input is the generated answer, and the output is what the user can see and hear.

[0401] Step 7:

[0402] (Collecting and processing feedback)

[0403] The user (store clerk) inputs feedback on the provided answer through the smart glasses. The smart glasses send this feedback to a cloud server. The server stores the feedback in a database and uses it to train the generative AI model. The input is the feedback data, and the output is the feedback stored in the database and updates to the generative AI model.

[0404] 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.

[0405] 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.

[0406] 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.

[0407] [Second embodiment]

[0408] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0409] 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.

[0410] 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).

[0411] 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.

[0412] 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.

[0413] 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).

[0414] 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.

[0415] 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.

[0416] 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.

[0417] 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.

[0418] 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.

[0419] 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."

[0420] The present invention is a learning support system in which a user (a student or a teacher) inputs a question and an answer to the question is provided using a generative AI model. Specific embodiments of the present invention are described below.

[0421] 1. User inputs a question

[0422] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0423] 2. Submit your question

[0424] The device has a function to send questions entered by the user to the server, and the entered data is sent to the server via a specific API endpoint.

[0425] 3. Receiving and analyzing questions

[0426] The server receives questions sent from the user's device, analyzes the data, and passes it to the generative AI model. Specifically, it converts the input text data into a format that the AI ​​model can understand.

[0427] 4. Answer Generation

[0428] The server then passes the analyzed question data to a generative AI model, which processes it to generate the optimal answer to the question. This generative AI model uses natural language processing technology and is capable of generating advanced answers based on past data and learning results.

[0429] 5. Returning the Response

[0430] The server then sends the generated answer back to the user's device, which may include, for example, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0431] 6. View Answers

[0432] The terminal displays the answer returned from the server to the user, allowing the user to confirm an immediate and appropriate answer to the question.

[0433] 7. Providing Feedback

[0434] Users can provide ratings and opinions on generated answers based on such things as whether the answer was helpful or easy to understand.

[0435] 8. Sending and Saving Feedback

[0436] The device sends user feedback to the server, which stores it in a database and uses it to train and improve the generative AI model, allowing the system to continuously improve its accuracy and quality.

[0437] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." The user can review this answer and provide a high rating if they find it specific and easy to understand. This feedback is stored in the system and used to generate future answers.

[0438] In this way, the present invention is a system that allows users to obtain immediate and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning.

[0439] The processing flow will be explained below.

[0440] Program processing flow

[0441] The specific processing flow of this system will be explained below step by step.

[0442] Step 1:

[0443] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0444] Step 2:

[0445] The device sends the data of the question entered by the user to the server using a dedicated API endpoint, and the entered question is sent in text data format.

[0446] Step 3:

[0447] The server receives questions sent from the user's device. The received question data is preprocessed for analysis and prepared for passing to the generative AI model. Specifically, the text data is tokenized and converted into a format that is easy for the AI ​​model to understand.

[0448] Step 4:

[0449] The server inputs the preprocessed question data into a generative AI model, which generates the optimal answer. This generative AI model uses natural language processing technology to generate answers based on past data and learning results.

[0450] Step 5:

[0451] The server then sends the generated answer back to the user's device, such as, "Photosynthesis is a process primarily performed by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0452] Step 6:

[0453] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers and deepen their understanding of the learning content.

[0454] Step 7:

[0455] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[0456] Step 8:

[0457] The terminal transmits the feedback from the user to the server, and the feedback data is transmitted in the form of ratings and comments.

[0458] Step 9:

[0459] The server stores the received feedback in a database. This feedback data is used to train and improve the generative AI model. Accumulating feedback data improves the accuracy and quality of the entire system.

[0460] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, maximizing the effectiveness of learning support. It also reduces the burden on teachers and provides learning support tailored to each student.

[0461] Example 1

[0462] 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."

[0463] In modern learning support systems, it is not easy to provide immediate and appropriate answers to questions submitted by users. Furthermore, improving the quality of generated answers and learning models by utilizing subsequent feedback are also important issues. Therefore, a system that can do this efficiently is needed.

[0464] 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.

[0465] In this invention, the server includes means for inputting a question through a user's terminal, means for transmitting the question to the server, means for receiving and analyzing the question at the server and passing it as a prompt to the generative AI model, means for generating an answer to the question using the generative AI model, means for returning the generated answer to the user's terminal, means for displaying the generated answer on the user's terminal, means for receiving feedback from the user, means for transmitting and storing the feedback to the server, and means for using the feedback to train the generative AI model. This allows users to receive instant, high-quality answers to their questions, and the system is continuously improved based on the feedback.

[0466] A "user's terminal" is an electronic device, such as a PC or smartphone, that a user uses to input information.

[0467] A "question" is a textual inquiry that a user enters into the system.

[0468] The "server" is a central processing unit that receives and analyzes instructions sent from the user's device and passes them to the generative AI model.

[0469] A "generative AI model" refers to an artificial intelligence algorithm that uses natural language processing technology to generate optimal answers to input questions.

[0470] A "prompt" is a pre-processed textual instruction that is passed to a generative AI model.

[0471] An "answer" is a text response generated by a generative AI model in response to a user's question.

[0472] "Feedback" refers to the ratings and opinions that users provide on generated answers.

[0473] A "database" is a data storage system for storing and managing data such as feedback.

[0474] "Natural language processing technology" is a computational technology for analyzing, understanding, and generating language that humans use on a daily basis.

[0475] This invention is a learning support system that uses a generative AI model to provide appropriate answers to questions entered by a user. This system consists of multiple components, including a user terminal, a server, and a generative AI model.

[0476] 1. User inputs a question

[0477] Users use a device such as a PC or smartphone to enter questions into a dedicated input interface. Specifically, they open a web browser, access the system's web page, and enter a question into the input form, such as "Please tell me about the process of photosynthesis."

[0478] 2. Submit your question

[0479] The device sends the question entered by the user to the server as an HTTP POST request via an API endpoint. The data is transferred in JSON format, for example:

[0480] json

[0481] {

[0482] "question": "What is the process of photosynthesis?"

[0483] }

[0484] 3. Receiving and analyzing questions

[0485] The server receives the HTTP POST request sent from the device and analyzes the question data contained therein. This analysis includes normalizing the text data and removing unnecessary whitespace and special characters. This converts the text data into a format that the generative AI model can understand.

[0486] 4. Answer Generation

[0487] The server then passes the analyzed question data to a generative AI model as a prompt to generate the optimal answer. This generative AI model uses natural language processing technology, such as OpenAI's GPT-3, which runs on the cloud. The prompt is entered in the following format:

[0488] text

[0489] "Tell me about the process of photosynthesis."

[0490] Generative AI models take into account past data and context to generate answers like:

[0491] text

[0492] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0493] 5. Returning and Displaying Answers

[0494] The server returns the generated answer in JSON format to the user's device as an HTTP response. The device displays the answer received from the server on the screen and tells the user, for example, as follows:

[0495] text

[0496] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0497] 6. Providing and Submitting Feedback

[0498] The user inputs their rating and opinion for the generated answer. For example, they select a rating such as "very helpful" in the rating form. The device sends this rating to the server as an HTTP POST request. An example of the data to be sent is as follows:

[0499] json

[0500] {

[0501] "question": "Explain the process of photosynthesis",

[0502] "answer": "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen.",

[0503] "feedback": "Very helpful"

[0504] }

[0505] 7. Storage and Use of Feedback

[0506] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance and answer accuracy.

[0507] In this way, the system of the present invention allows users to receive instant, high-quality answers to their questions and is continuously improved based on feedback, thereby maximizing the effectiveness of learning support, reducing the burden on teachers, and ultimately improving the quality of learning.

[0508] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0509] Step 1:

[0510] A user inputs a question into a dedicated input interface using a device such as a PC or smartphone. For example, the input question might be text such as "Please tell me about the process of photosynthesis." This input is completed when the user enters the text into an HTML form element and presses the submit button.

[0511] Input: User-provided question text

[0512] Output: Question text in terminal input form

[0513] Step 2:

[0514] The device sends the question entered by the user to the server by sending an HTTP POST request to the API endpoint. The question text is encoded in JSON format and sent.

[0515] Input: User question text

[0516] Output: JSON formatted data sent to the server

[0517] Step 3:

[0518] The server receives an HTTP POST request sent from the device, which contains the user's question. At this stage, the server extracts the question text from the request and normalizes it, removing unnecessary whitespace and special characters.

[0519] Input: JSON format data received from the terminal

[0520] Output: Normalized question text

[0521] Step 4:

[0522] The server passes the normalized question text as a prompt to a generative AI model, which uses natural language processing techniques, for example. The server sends the prompt to the generative AI model, which then generates an answer.

[0523] Input: Normalized question text

[0524] Output: Generated answer text

[0525] Step 5:

[0526] The server packages the answer received from the generative AI model in JSON format and returns it to the user's device as an HTTP response, which includes the generated answer text.

[0527] Input: Answer text from the generative AI model

[0528] Output: JSON formatted response data

[0529] Step 6:

[0530] The terminal processes the HTTP response received from the server and displays the generated answer to the user in HTML format. Using JavaScript, the answer text is inserted into the specified DOM element on the browser.

[0531] Input: JSON format response data from the server

[0532] Output: Answer text displayed in the browser

[0533] Step 7:

[0534] The user can enter feedback on the displayed answer by using the evaluation form to select a rating such as "helpful" or "easy to understand" and enter comments if necessary.

[0535] Input: User feedback on the generated answer

[0536] Output: Feedback text

[0537] Step 8:

[0538] The device sends the user-entered feedback to the server, again using an HTTP POST request, with the feedback encoded in JSON format.

[0539] Input: User feedback text

[0540] Output: Feedback data sent to the server in JSON format.

[0541] Step 9:

[0542] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance.

[0543] Input: Feedback data received from the device in JSON format

[0544] Output: Feedback data stored in a database

[0545] (Application example 1)

[0546] 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."

[0547] In traditional brick-and-mortar stores, when customers wanted detailed information about a product or a guided tour of the store, it was difficult to provide them with the information quickly and accurately. Furthermore, because the store required the assistance of an employee, responses were often delayed during busy periods, resulting in lower customer satisfaction. Furthermore, the quality and speed of responses to customer questions depended on the experience and knowledge of the employee, which meant that consistent service could be inconsistent. To solve these issues, a system was needed that would allow customers to input questions and receive instant answers.

[0548] 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.

[0549] In this invention, the server includes a means for accepting user input, a means for analyzing the input and passing it to the generative AI model, and a means for returning the generated answer to the user terminal, thereby enabling quick and accurate answers to questions about product explanations, store guides, and recipe suggestions in a physical store to be provided.

[0550] The "means for accepting user input" is a function that allows a user to input questions or requests to the terminal via a dedicated input interface.

[0551] The "means for analyzing input and passing it to the generative artificial intelligence model" is a processing function for analyzing data input by a user, converting it into a format that can be understood by the artificial intelligence model, and passing it on.

[0552] "Means for generating answers to user input using a generative artificial intelligence model" refers to algorithms and techniques that allow the artificial intelligence model to generate optimal answers based on analyzed input data.

[0553] "Means for returning the generated answer to the user terminal" refers to communication and display technology for transferring and displaying the answer generated by the artificial intelligence model to the user's terminal.

[0554] "Means for receiving feedback from users and using it to train the generative AI model" refers to data collection and learning technology that collects evaluations and opinions on answers provided by users and uses them to improve and enhance the accuracy of the AI ​​model.

[0555] "Means of providing answers to questions regarding product descriptions, store guidance, and recipe suggestions within a physical store" refers to functions and technologies that allow customers to ask questions regarding product and store information and recipe suggestions in a physical store, and for an AI model to provide immediate and appropriate answers.

[0556] "Means for storing data in a database" refers to the storage systems and technologies used to organize and securely store collected data and feedback.

[0557] "Using natural language processing technology" refers to the process of analyzing text entered by a user and generating an answer using algorithms and technologies for understanding and generating human language.

[0558] A specific embodiment of the present invention will be described in detail below. This invention is a system that allows users (customers) to ask questions about products, store information, and recipe suggestions in a physical store and receive automatic answers.

[0559] System Configuration

[0560] Hardware

[0561] The server uses a virtual machine on the cloud (e.g., AWS EC2, Google Cloud VM), and the user's device is a smartphone (iOS / Android) or PC.

[0562] software

[0563] The programming language used is Python.

[0564] Use Flask as a web framework.

[0565] To analyze questions and generate answers, we use generative AI models provided by OpenAI (e.g., GPT-3).

[0566] Detailed System Description

[0567] User Input

[0568] Users input questions using a dedicated input interface on their smartphone or PC, such as "Where is this wine produced?" or "Please tell me some recipes using this cheese."

[0569] Submit a Question

[0570] The device sends the entered question to a server in the cloud via a specific API endpoint.

[0571] Receiving and parsing questions

[0572] The server receives questions sent by users, analyzes the data, and passes it to a generative AI model (e.g., GPT-3). Specifically, it converts the text data into a format that the AI ​​model can understand.

[0573] Generate answers

[0574] The server uses the analyzed question data to input the generative AI model and generate the optimal answer, such as "This wine is produced in the Bordeaux region of France."

[0575] Returning the answer

[0576] The server returns the generated answer to the user terminal, which displays the returned answer on the screen for the user to confirm.

[0577] Providing and storing feedback

[0578] Users can provide ratings and opinions on generated answers based on their helpfulness and ease of understanding. Feedback is sent to the server and stored in a database. This information is used to continuously learn and improve the generative AI model.

[0579] Specific examples

[0580] For example, if a user types "Where is this wine from?" the system generates the following prompt:

[0581] Prompt Sentence Examples

[0582] Prompt: "Q: Where is this wine from?\nA:"

[0583] The server passes this prompt to the generative AI model, which then generates the answer, "This wine is produced in the Bordeaux region of France." The user receives this information through their device and provides a high rating if they are satisfied with the answer.

[0584] The above is a detailed description of an embodiment of the present invention. The present invention is a system that can improve convenience for customers in a physical store and reduce the burden on employees.

[0585] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0586] Step 1:

[0587] The user inputs a question using the input interface on the terminal. For example, the user can input the question, "Where is this wine produced?" The input data is in text format, and the terminal sends it to the server.

[0588] Step 2:

[0589] The device sends the entered question to a server on the cloud via a specific API endpoint. The sent data is in text format. The server receives this data.

[0590] Step 3:

[0591] The server receives the question sent by the user and analyzes the text data. The analyzed data is converted into a format that can be understood by a generative AI model (e.g., GPT-3). In this step, a prompt sentence is generated. For example, the format is "Prompt: "Q: Where is this wine produced?\nA:"".

[0592] Step 4:

[0593] The server inputs the generated prompt sentence into the generative AI model, which then generates the optimal answer based on the prompt. Data processing involves converting the input text into vector format and applying natural language processing algorithms to generate the answer.

[0594] Step 5:

[0595] The server receives the answer generated by the generative AI model and prepares it to be sent back. The generated answer is in text format. For example, the answer generated might be, "This wine is produced in the Bordeaux region of France."

[0596] Step 6:

[0597] The server sends the generated answer back to the user's terminal, which displays the received answer on the screen, allowing the user to check the answer to the question.

[0598] Step 7:

[0599] Users provide ratings and opinions on the generated answers, such as whether the answer was helpful or easy to understand. The rating data is entered in text format or multiple-choice format.

[0600] Step 8:

[0601] The device sends user feedback to a server, which receives it and stores it in a database. The stored evaluation data is used to train and improve the generative AI model in the future.

[0602] The above is the specific processing flow of the program that realizes this application example. This flow makes it possible to provide quick and accurate answers to customer questions in a physical store.

[0603] 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.

[0604] The present invention is a learning support system in which users (students or teachers) input questions and answers are provided using a generative AI model. Furthermore, the present invention achieves more accurate learning support by combining an emotion engine that recognizes the user's emotions. Specific embodiments of the present invention are described below.

[0605] 1. User inputs a question

[0606] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0607] 2. Starting Emotion Recognition

[0608] The device recognizes the user's emotions by using facial expressions and voice data when the user inputs a question. This emotion recognition is performed using an emotion engine.

[0609] 3. Sending questions and emotion data

[0610] The device sends the question entered by the user and the accompanying emotion data to the server. The entered data is sent as question text data and emotion data from the emotion engine.

[0611] 4. Receiving and analyzing questions

[0612] The server receives and analyzes questions and emotion data sent from the user's device. The question data is preprocessed and passed to the generative AI model, and the emotion data is used to generate and adjust the answer.

[0613] 5. Generate and refine answers

[0614] The server uses a generative AI model to generate optimal answers to questions. Furthermore, it adjusts the tone and content of the answer based on the user's emotions, based on emotional data from the emotion engine. For example, if the user looks anxious, a more friendly and polite answer will be generated.

[0615] 6. Returning the Response

[0616] The server then sends a tailored response back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Easy to understand, right?"

[0617] 7. View Answers

[0618] The terminal displays the answers returned from the server to the user, who can then check the adjusted answers and deepen their understanding of the learning content.

[0619] 8. Providing Feedback

[0620] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[0621] 9. Sending and Saving Feedback

[0622] The device sends user feedback to the server, which stores it in a database and uses it to subsequently train and improve the generative AI model and emotion engine, thereby improving the accuracy and quality of the entire system.

[0623] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with, "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." Furthermore, if the user looks anxious, the system will include additional explanation such as, "It's okay, let's understand it well." The user will confirm this response and appreciate the detailed and easy-to-understand explanation. In this way, by utilizing the emotion engine, it is possible to provide customized learning support according to the user's emotions.

[0624] In this way, the present invention is a system that allows users to receive prompt and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning. Furthermore, by using an emotion engine, support that takes into consideration the user's emotions can be realized, providing even greater satisfaction and understanding.

[0625] The processing flow will be explained below.

[0626] Program processing flow

[0627] Step 1:

[0628] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0629] Step 2:

[0630] The device receives the question entered by the user as text data and simultaneously acquires emotion data from the user's facial expressions and voice. The emotion data is used to identify the user's emotion using video and audio analysis approaches.

[0631] Step 3:

[0632] The device transmits the acquired question data and emotion data to the server using a secure communication protocol.

[0633] Step 4:

[0634] The server receives question data and emotion data sent from the user's device. After receiving the data, it performs preprocessing such as tokenizing the text data, and prepares it for passing to the generative AI model and emotion engine.

[0635] Step 5:

[0636] The server uses a generative artificial intelligence model to generate optimal answers based on the question data. The generated answers are based on the underlying learning content.

[0637] Step 6:

[0638] The server uses an emotion engine to tailor the generated responses, specifically adjusting the tone and expression of the responses based on the user's emotion data. For example, if the user expresses anxiety, the responses will be more kind and polite.

[0639] Step 7:

[0640] The server then sends tailored answers back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Was this easy to understand?"

[0641] Step 8:

[0642] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers to deepen their understanding of the learning content.

[0643] Step 9:

[0644] Users provide ratings and opinions (feedback) on the generated answers, such as whether the answers were helpful, easy to understand, or whether they made them feel a certain emotion.

[0645] Step 10:

[0646] The device sends user feedback and emotion data to the server, which will be used to improve the system in the future.

[0647] Step 11:

[0648] The server stores the received feedback and emotion data in a database. This data is used to train and improve the generative AI model and emotion engine. Accumulating data improves the accuracy and quality of the system.

[0649] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, and by using an emotion engine, it provides personalized learning support according to the user's emotions. Furthermore, feedback data can be used to continuously improve the system.

[0650] Example 2

[0651] 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."

[0652] Conventional learning support systems can use artificial intelligence (AI) to provide answers to user questions, but do not generate answers that take the user's emotions into account. This makes it difficult to address the anxiety and confusion users feel while learning, which can reduce the quality and effectiveness of learning. Furthermore, there are also insufficient mechanisms for appropriately accumulating user feedback and reflecting it in subsequent system improvements. The present invention aims to solve these problems.

[0653] 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 means for accepting user input, means for recognizing the user's emotion, means for analyzing the input and emotion data and passing it to a generative AI model, means for generating a response to the user's input using the generative AI model and adjusting the response based on the user's emotion, means for returning the generated response to the user terminal, and means for receiving feedback from the user and using it for training the generative AI model and the emotion recognition means. As a result, by recognizing and responding to the user's emotion, it is possible to provide more appropriate and effective learning support and improve the accuracy and quality of the entire system based on the feedback.

[0654] The "means for accepting user input" refers to a means by which a user inputs questions or information related to learning in the form of text, voice, or the like.

[0655] The "means for recognizing the user's emotions" is a means for analyzing the user's facial expressions and voice and identifying the emotions the user is feeling.

[0656] The "means for analyzing the input and emotion data and passing it to the generative AI model" refers to a means for analyzing the user's question text and voice, as well as the recognized emotion data, and passing it to the generative AI model.

[0657] "Means for using a generative artificial intelligence model to generate a response to a user's input and adjusting the response based on the user's emotions" refers to means for using a generative AI model to generate an optimal response to a user's question and further adjusting the content and tone of the response depending on the user's emotions.

[0658] The "means for returning the generated answer to the user terminal" refers to a means for transmitting the generated and adjusted answer to the user's terminal.

[0659] "Means for receiving feedback from users and using it to train the generative artificial intelligence model and emotion recognition means" refers to means for receiving evaluations and opinions on answers provided by users and using them to improve the performance of the generative artificial intelligence model and emotion recognition engine.

[0660] "Means for storing in a database" means a means for permanently recording user feedback and other relevant data for later analysis and system improvement.

[0661] "Using natural language processing technology" means using technology that applies natural language processing algorithms to the analysis and generation of text data.

[0662] MODE FOR CARRYING OUT THE INVENTION

[0663] The present invention is a system that uses a generative AI model to provide answers to questions about learning entered by a user, and further combines it with an emotion engine to enable customized learning support according to emotions. The following describes in detail the embodiments for implementing this system.

[0664] 1. System Configuration

[0665] This system is broadly composed of a user's device, a server, and a cloud service that includes an emotion engine and a generative AI model.

[0666] User's device

[0667] Users access a dedicated learning support input interface using devices such as PCs or smartphones. This interface is web browser-based software (e.g., implemented using React.js) and is intuitive to operate.

[0668] server

[0669] The server is the central part that receives input from the user and performs the necessary analysis. It is powered by Node.js and Express.js and consists of the following components:

[0670] Input receiving component: Receives question text and sentiment data from the user.

[0671] Preprocessing component: Performs preprocessing such as tokenizing the question text and removing unnecessary symbols.

[0672] Generative AI model invocation component: Invokes a generative AI model, such as OpenAI's GPT-4, to generate answers to questions.

[0673] Emotion adjustment component: Adjusts the generated answers based on the user's emotional data.

[0674] Feedback collection component: receives feedback from users and stores it in a database.

[0675] Emotion Engine

[0676] The emotion engine analyzes the user's facial expressions and voice to identify the emotions they are feeling, using the Emotion API from Microsoft Azure Cognitive Services and the Cloud Speech-to-Text API from Google Cloud.

[0677] Description of software and hardware used

[0678] Webcam and microphone: Hardware that captures a user's facial expressions and records their voice.

[0679] Microsoft Azure Cognitive Services Emotion API: Software for analyzing facial expression data and recognizing emotions.

[0680] Google Cloud Speech-to-Text API: Software for analyzing voice data and recognizing emotions.

[0681] OpenAI GPT-4: A generative AI model that generates answers to questions.

[0682] Node.js and Express.js: Software for performing server-side processing.

[0683] Example of a system

[0684] As a concrete example, if a user types the question "Tell me about the process of photosynthesis," the system will do the following:

[0685] 1. The user types "Please tell me about the process of photosynthesis" into the PC's input interface.

[0686] 2. The device uses a webcam and microphone to capture the user's facial expressions and voice and sends the data to the emotion engine.

[0687] 3. The emotion engine analyzes the captured data and identifies the user's emotion as "anxiety."

[0688] 4. The device sends the entered question and emotion data to the server.

[0689] 5. The server preprocesses the received data and passes it to a generative AI model to generate an answer.

[0690] 6. After the answer is generated, the server takes into account the sentiment data and adds additional encouragement to the answer, such as "It's okay, you'll understand."

[0691] 7. The server sends the tailored response back to the user's terminal.

[0692] 8. The terminal displays the answer for the user to confirm.

[0693] 9. The user rates the answer and provides feedback on whether it was helpful.

[0694] 10. The device sends the feedback to the server, which stores it in a database.

[0695] Prompt Sentence Examples

[0696] Examples of prompts include:

[0697] "Please explain the process of photosynthesis in detail. The user looks anxious."

[0698] In this way, the present invention can provide customized learning assistance that takes into account the user's emotions, improving the quality of learning.

[0699] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0700] Step 1: User enters question

[0701] The user inputs a question related to the study into the input interface of the terminal. The input question text is sent to the interface. For example, by inputting "Please tell me about the process of photosynthesis," the text data is sent to the terminal.

[0702] Step 2: Start Recognizing Emotions

[0703] The device captures the user's facial expressions with a webcam and records their voice with a microphone. The captured data (image data and voice data) is sent to an emotion engine. Specific emotion engines used here include the Emotion API and Cloud Speech-to-Text API. The emotion engine then analyzes the input data and generates emotion data.

[0704] Step 3: Submit your question and sentiment data

[0705] The device sends the entered question text and analyzed emotion data in JSON format to the server. The sent data includes, for example, the following format:

[0706] json

[0707] {

[0708] "question": "Please explain the process of photosynthesis.",

[0709] "emotion": {

[0710] "type": "anxious",

[0711] "confidence": 0.85

[0712] }

[0713] }

[0714] This causes the data to be sent to the server for further processing.

[0715] Step 4: Receive and analyze question and sentiment data

[0716] The server receives the data sent from the device. The received data is divided into question text and emotion data. The server tokenizes the question text and performs preprocessing to remove unnecessary symbols, generating data for analysis. The tokenized question text is then sent to the subsequent generative AI model.

[0717] Step 5: Generate and refine answers

[0718] The server uses a generative AI model to generate answers to questions. Specifically, it invokes OpenAI's GPT-4 model and retrieves the generated answer. For example, it might generate an answer like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen." It then adjusts the tone and content of the answer based on emotional data. If the user shows signs of anxiety, it might add a supporting sentence like, "It's okay, let's understand it."

[0719] Step 6: Submit your response

[0720] The server encodes the adjusted answer in JSON format and returns it to the user's device as an HTTP response, which sends the adjusted answer to the user's device.

[0721] Step 7: View your answers

[0722] The device decodes the response data received from the server and displays it to the user. A React.js component is used for display, allowing the user to view the adjusted response in their browser. For example, the device might display something like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen. It's okay, just make sure you understand it."

[0723] Step 8: Provide feedback

[0724] Users can provide feedback on the displayed answers through a special evaluation form, which includes options such as "helpful" or "confusing."

[0725] Step 9: Submit and save your feedback

[0726] The device encodes the user-provided feedback in JSON format and sends it to the server, which then receives the feedback and stores it in a database. This allows the feedback data to be accumulated and used for future training of the generative AI model and emotion engine.

[0727] (Application example 2)

[0728] 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."

[0729] Conventional learning support systems and customer service systems only provide uniform answers to questions entered by users (students or store clerks), and have the problem of not responding appropriately to the user's emotional state. This can result in insufficient learning benefits or customer service experience, and can lead to a decrease in user satisfaction. Therefore, there is a need for a system that can recognize the user's emotional state in real time and provide appropriate answers based on that.

[0730] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting user input, means for analyzing the input and passing it to a generative AI model, means for generating a response to the user input using the generative AI model, means for returning the generated response to the user terminal, means for receiving feedback from the user and using it for training the generative AI model, means for acquiring user emotion data using emotion recognition means built into the smart device, and means for analyzing the emotion data and adjusting the tone and content of the response. This makes it possible to generate customized responses according to the user's emotional state, thereby improving the quality of learning support and customer service support.

[0731] "User" refers to anyone who uses the system, and specifically includes learners and store clerks who serve customers.

[0732] "Means for accepting input" refers to the interface or device through which a user enters questions or instructions.

[0733] "Means for analyzing and passing to the generative artificial intelligence model" refers to the process for analyzing input data, converting it into an appropriate format, and passing it to the generative artificial intelligence model.

[0734] A "generative artificial intelligence model" refers to an AI model that uses natural language processing technology to generate answers to user input.

[0735] "Means for generating an answer" refers to a function that uses a generative artificial intelligence model to generate the optimal answer to a user's question.

[0736] "Means for returning the generated answer to the user device" refers to the function of sending the answer generated by AI to the device used by the user.

[0737] "Means of receiving feedback and using it for learning" refers to the function of collecting user evaluations and opinions and using them to improve the performance of the generative AI model.

[0738] A "smart device" is a device equipped with communication and computing capabilities, and specifically includes smart glasses, smartphones, and head-mounted displays.

[0739] "Emotion recognition means" refers to technology that uses a camera or microphone installed on the device to recognize emotions from the user's facial expressions and voice.

[0740] "Emotion data" refers to data that indicates the user's emotional state, and includes, for example, emotional states such as joy, anxiety, and anger.

[0741] "Means for adjusting the tone and content of responses" refers to a function that appropriately adjusts the expression and content of generated responses based on emotional data.

[0742] The present invention is embodied as a customer service support system for brick-and-mortar stores. Through an application using smart glasses called "Smart Tutor for Retail," the system provides quick and accurate answers to customer questions. An embodiment of this system will be described in detail below.

[0743] 1. User inputs a question

[0744] The user (store clerk) wears smart glasses and receives questions from customers as voice. For example, if a customer asks, "Please tell me more about this product," the smart glasses' high-performance microphone captures the voice.

[0745] 2. Voice Recognition

[0746] The captured voice data is converted into text by the voice recognition software in the smart glasses, using the Google Cloud Speech-to-Text API.

[0747] 3. Starting Emotion Recognition

[0748] The smart glasses use a built-in camera and microphone to capture the user's facial expressions and voice, thereby collecting emotional data. Emotion recognition is performed using Microsoft Azure Face API and voice analysis software.

[0749] 4. Sending questions and emotion data

[0750] The smart glasses send the converted text data and emotion data to a cloud server, where they are stored via GCP (Google Cloud Platform).

[0751] 5. Parsing Questions and Generating Answers

[0752] The server analyzes the question text using natural language processing (NLP) techniques to generate appropriate answers. The generative AI model used here is OpenAI's GPT-4. The generated answers are adjusted in tone and content based on emotional data.

[0753] 6. Returning and Displaying Tailored Responses

[0754] The server sends the adjusted answer back to the smart glasses, which display the answer on a display for the store clerk to see and also provide an audible response.

[0755] Examples and prompts

[0756] As a concrete example, if a customer asks, "Please tell me more about this product," the process will be as follows:

[0757] Example 1:

[0758] Customer: "Please give me a detailed description of this product."

[0759] Store clerk (through smart glasses): [Camera and microphone capture customer's question]

[0760] System: "This product is made using the latest technology and is particularly durable. For detailed characteristics, please see this URL."

[0761] Also, if the store clerk looks anxious, a more helpful and detailed explanation will be added.

[0762] Example prompt sentence:

[0763] User Question: "Can you give me a detailed description of this product?"

[0764] Prompt to generative AI model: "A customer has asked for a detailed description of a product. Please provide detailed information about the product's features, benefits, and usage, but keep it concise so that the sales associate can easily understand."

[0765] The key feature of this invention is that it recognizes the user's emotions in real time and provides customized responses based on those emotions. This improves the quality of customer service support and increases user satisfaction. In addition, the generative AI model is trained based on the collected feedback, improving the performance of the entire system.

[0766] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0767] Step 1:

[0768] (User inputs question)

[0769] The user (store clerk) wears smart glasses and receives voice questions from customers. For example, the input is voice data such as "Please tell me more about this product."

[0770] Step 2:

[0771] (Voice Recognition)

[0772] The microphone built into the smart glasses captures the received voice data, and then uses voice recognition software (Google Cloud Speech-to-Text API) to convert the voice data into text data. The input is voice data, and the output is text data.

[0773] Step 3:

[0774] (Start of emotion recognition)

[0775] The smart glasses use a built-in camera and microphone to capture the user's facial and voice data, and use the Microsoft Azure Face API to collect the user's emotion data from this data. The input is the captured facial and voice data, and the output is emotion data.

[0776] Step 4:

[0777] (Submitting questions and emotion data)

[0778] The smart glasses send text data and emotion data to a cloud server (GCP). The input is text data and emotion data, and the output is that this data is sent to the cloud server.

[0779] Step 5:

[0780] (Question analysis and answer generation)

[0781] The server analyzes the text data using natural language processing (NLP) techniques. The analyzed question is passed to a generative AI model (e.g., OpenAI's GPT-4) to generate the optimal answer. The tone and content of the generated answer are adjusted based on the emotional data. The input is the text data and emotional data, and the output is the generated answer.

[0782] Step 6:

[0783] (Return and display of adjusted answers)

[0784] The server sends the adjusted answer back to the smart glasses, which display the answer on their display and also speak it aloud. The input is the generated answer, and the output is what the user can see and hear.

[0785] Step 7:

[0786] (Collecting and processing feedback)

[0787] The user (store clerk) inputs feedback on the provided answer through the smart glasses. The smart glasses send this feedback to a cloud server. The server stores the feedback in a database and uses it to train the generative AI model. The input is the feedback data, and the output is the feedback stored in the database and updates to the generative AI model.

[0788] 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.

[0789] 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.

[0790] 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.

[0791] [Third embodiment]

[0792] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0793] 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.

[0794] 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).

[0795] 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.

[0796] 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.

[0797] 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).

[0798] 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.

[0799] 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.

[0800] 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.

[0801] 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.

[0802] 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.

[0803] 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."

[0804] The present invention is a learning support system in which a user (a student or a teacher) inputs a question and an answer to the question is provided using a generative AI model. Specific embodiments of the present invention are described below.

[0805] 1. User inputs a question

[0806] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0807] 2. Submit your question

[0808] The device has a function to send questions entered by the user to the server, and the entered data is sent to the server via a specific API endpoint.

[0809] 3. Receiving and analyzing questions

[0810] The server receives questions sent from the user's device, analyzes the data, and passes it to the generative AI model. Specifically, it converts the input text data into a format that the AI ​​model can understand.

[0811] 4. Answer Generation

[0812] The server then passes the analyzed question data to a generative AI model, which processes it to generate the optimal answer to the question. This generative AI model uses natural language processing technology and is capable of generating advanced answers based on past data and learning results.

[0813] 5. Returning the Response

[0814] The server then sends the generated answer back to the user's device, which may include, for example, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0815] 6. View Answers

[0816] The terminal displays the answer returned from the server to the user, allowing the user to confirm an immediate and appropriate answer to the question.

[0817] 7. Providing Feedback

[0818] Users can provide ratings and opinions on generated answers based on such things as whether the answer was helpful or easy to understand.

[0819] 8. Sending and Saving Feedback

[0820] The device sends user feedback to the server, which stores it in a database and uses it to train and improve the generative AI model, allowing the system to continuously improve its accuracy and quality.

[0821] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." The user can review this answer and provide a high rating if they find it specific and easy to understand. This feedback is stored in the system and used to generate future answers.

[0822] In this way, the present invention is a system that allows users to obtain immediate and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning.

[0823] The processing flow will be explained below.

[0824] Program processing flow

[0825] The specific processing flow of this system will be explained below step by step.

[0826] Step 1:

[0827] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0828] Step 2:

[0829] The device sends the data of the question entered by the user to the server using a dedicated API endpoint, and the entered question is sent in text data format.

[0830] Step 3:

[0831] The server receives questions sent from the user's device. The received question data is preprocessed for analysis and prepared for passing to the generative AI model. Specifically, the text data is tokenized and converted into a format that is easy for the AI ​​model to understand.

[0832] Step 4:

[0833] The server inputs the preprocessed question data into a generative AI model, which generates the optimal answer. This generative AI model uses natural language processing technology to generate answers based on past data and learning results.

[0834] Step 5:

[0835] The server then sends the generated answer back to the user's device, such as, "Photosynthesis is a process primarily performed by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0836] Step 6:

[0837] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers and deepen their understanding of the learning content.

[0838] Step 7:

[0839] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[0840] Step 8:

[0841] The terminal transmits the feedback from the user to the server, and the feedback data is transmitted in the form of ratings and comments.

[0842] Step 9:

[0843] The server stores the received feedback in a database. This feedback data is used to train and improve the generative AI model. Accumulating feedback data improves the accuracy and quality of the entire system.

[0844] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, maximizing the effectiveness of learning support. It also reduces the burden on teachers and provides learning support tailored to each student.

[0845] Example 1

[0846] 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."

[0847] In modern learning support systems, it is not easy to provide immediate and appropriate answers to questions submitted by users. Furthermore, improving the quality of generated answers and learning models by utilizing subsequent feedback are also important issues. Therefore, a system that can do this efficiently is needed.

[0848] 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.

[0849] In this invention, the server includes means for inputting a question through a user's terminal, means for transmitting the question to the server, means for receiving and analyzing the question at the server and passing it as a prompt to the generative AI model, means for generating an answer to the question using the generative AI model, means for returning the generated answer to the user's terminal, means for displaying the generated answer on the user's terminal, means for receiving feedback from the user, means for transmitting and storing the feedback to the server, and means for using the feedback to train the generative AI model. This allows users to receive instant, high-quality answers to their questions, and the system is continuously improved based on the feedback.

[0850] A "user's terminal" is an electronic device, such as a PC or smartphone, that a user uses to input information.

[0851] A "question" is a textual inquiry that a user enters into the system.

[0852] The "server" is a central processing unit that receives and analyzes instructions sent from the user's device and passes them to the generative AI model.

[0853] A "generative AI model" refers to an artificial intelligence algorithm that uses natural language processing technology to generate optimal answers to input questions.

[0854] A "prompt" is a pre-processed textual instruction that is passed to a generative AI model.

[0855] An "answer" is a text response generated by a generative AI model in response to a user's question.

[0856] "Feedback" refers to the ratings and opinions that users provide on generated answers.

[0857] A "database" is a data storage system for storing and managing data such as feedback.

[0858] "Natural language processing technology" is a computational technology for analyzing, understanding, and generating language that humans use on a daily basis.

[0859] This invention is a learning support system that uses a generative AI model to provide appropriate answers to questions entered by a user. This system consists of multiple components, including a user terminal, a server, and a generative AI model.

[0860] 1. User inputs a question

[0861] Users use a device such as a PC or smartphone to enter questions into a dedicated input interface. Specifically, they open a web browser, access the system's web page, and enter a question into the input form, such as "Please tell me about the process of photosynthesis."

[0862] 2. Submit your question

[0863] The device sends the question entered by the user to the server as an HTTP POST request via an API endpoint. The data is transferred in JSON format, for example:

[0864] json

[0865] {

[0866] "question": "What is the process of photosynthesis?"

[0867] }

[0868] 3. Receiving and analyzing questions

[0869] The server receives the HTTP POST request sent from the device and analyzes the question data contained therein. This analysis includes normalizing the text data and removing unnecessary whitespace and special characters. This converts the text data into a format that the generative AI model can understand.

[0870] 4. Answer Generation

[0871] The server then passes the analyzed question data to a generative AI model as a prompt to generate the optimal answer. This generative AI model uses natural language processing technology, such as OpenAI's GPT-3, which runs on the cloud. The prompt is entered in the following format:

[0872] text

[0873] "Tell me about the process of photosynthesis."

[0874] Generative AI models take into account past data and context to generate answers like:

[0875] text

[0876] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0877] 5. Returning and Displaying Answers

[0878] The server returns the generated answer in JSON format to the user's device as an HTTP response. The device displays the answer received from the server on the screen and tells the user, for example, as follows:

[0879] text

[0880] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[0881] 6. Providing and Submitting Feedback

[0882] The user inputs their rating and opinion for the generated answer. For example, they select a rating such as "very helpful" in the rating form. The device sends this rating to the server as an HTTP POST request. An example of the data to be sent is as follows:

[0883] json

[0884] {

[0885] "question": "Explain the process of photosynthesis",

[0886] "answer": "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen.",

[0887] "feedback": "Very helpful"

[0888] }

[0889] 7. Storage and Use of Feedback

[0890] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance and answer accuracy.

[0891] In this way, the system of the present invention allows users to receive instant, high-quality answers to their questions and is continuously improved based on feedback, thereby maximizing the effectiveness of learning support, reducing the burden on teachers, and ultimately improving the quality of learning.

[0892] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0893] Step 1:

[0894] A user inputs a question into a dedicated input interface using a device such as a PC or smartphone. For example, the input question might be text such as "Please tell me about the process of photosynthesis." This input is completed when the user enters the text into an HTML form element and presses the submit button.

[0895] Input: User-provided question text

[0896] Output: Question text in terminal input form

[0897] Step 2:

[0898] The device sends the question entered by the user to the server by sending an HTTP POST request to the API endpoint. The question text is encoded in JSON format and sent.

[0899] Input: User question text

[0900] Output: JSON formatted data sent to the server

[0901] Step 3:

[0902] The server receives an HTTP POST request sent from the device, which contains the user's question. At this stage, the server extracts the question text from the request and normalizes it, removing unnecessary whitespace and special characters.

[0903] Input: JSON format data received from the terminal

[0904] Output: Normalized question text

[0905] Step 4:

[0906] The server passes the normalized question text as a prompt to a generative AI model, which uses natural language processing techniques, for example. The server sends the prompt to the generative AI model, which then generates an answer.

[0907] Input: Normalized question text

[0908] Output: Generated answer text

[0909] Step 5:

[0910] The server packages the answer received from the generative AI model in JSON format and returns it to the user's device as an HTTP response, which includes the generated answer text.

[0911] Input: Answer text from the generative AI model

[0912] Output: JSON formatted response data

[0913] Step 6:

[0914] The terminal processes the HTTP response received from the server and displays the generated answer to the user in HTML format. Using JavaScript, the answer text is inserted into the specified DOM element on the browser.

[0915] Input: JSON format response data from the server

[0916] Output: Answer text displayed in the browser

[0917] Step 7:

[0918] The user can enter feedback on the displayed answer by using the evaluation form to select a rating such as "helpful" or "easy to understand" and enter comments if necessary.

[0919] Input: User feedback on the generated answer

[0920] Output: Feedback text

[0921] Step 8:

[0922] The device sends the user-entered feedback to the server, again using an HTTP POST request, with the feedback encoded in JSON format.

[0923] Input: User feedback text

[0924] Output: Feedback data sent to the server in JSON format.

[0925] Step 9:

[0926] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance.

[0927] Input: Feedback data received from the device in JSON format

[0928] Output: Feedback data stored in a database

[0929] (Application example 1)

[0930] 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."

[0931] In traditional brick-and-mortar stores, when customers wanted detailed information about a product or a guided tour of the store, it was difficult to provide them with the information quickly and accurately. Furthermore, because the store required the assistance of an employee, responses were often delayed during busy periods, resulting in lower customer satisfaction. Furthermore, the quality and speed of responses to customer questions depended on the experience and knowledge of the employee, which meant that consistent service could be inconsistent. To solve these issues, a system was needed that would allow customers to input questions and receive instant answers.

[0932] 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.

[0933] In this invention, the server includes a means for accepting user input, a means for analyzing the input and passing it to the generative AI model, and a means for returning the generated answer to the user terminal, thereby enabling quick and accurate answers to questions about product explanations, store guides, and recipe suggestions in a physical store to be provided.

[0934] The "means for accepting user input" is a function that allows a user to input questions or requests to the terminal via a dedicated input interface.

[0935] The "means for analyzing input and passing it to the generative artificial intelligence model" is a processing function for analyzing data input by a user, converting it into a format that can be understood by the artificial intelligence model, and passing it on.

[0936] "Means for generating answers to user input using a generative artificial intelligence model" refers to algorithms and techniques that allow the artificial intelligence model to generate optimal answers based on analyzed input data.

[0937] "Means for returning the generated answer to the user terminal" refers to communication and display technology for transferring and displaying the answer generated by the artificial intelligence model to the user's terminal.

[0938] "Means for receiving feedback from users and using it to train the generative AI model" refers to data collection and learning technology that collects evaluations and opinions on answers provided by users and uses them to improve and enhance the accuracy of the AI ​​model.

[0939] "Means of providing answers to questions regarding product descriptions, store guidance, and recipe suggestions within a physical store" refers to functions and technologies that allow customers to ask questions regarding product and store information and recipe suggestions in a physical store, and for an AI model to provide immediate and appropriate answers.

[0940] "Means for storing data in a database" refers to the storage systems and technologies used to organize and securely store collected data and feedback.

[0941] "Using natural language processing technology" refers to the process of analyzing text entered by a user and generating an answer using algorithms and technologies for understanding and generating human language.

[0942] A specific embodiment of the present invention will be described in detail below. This invention is a system that allows users (customers) to ask questions about products, store information, and recipe suggestions in a physical store and receive automatic answers.

[0943] System Configuration

[0944] Hardware

[0945] The server uses a virtual machine on the cloud (e.g., AWS EC2, Google Cloud VM), and the user's device is a smartphone (iOS / Android) or PC.

[0946] software

[0947] The programming language used is Python.

[0948] Use Flask as a web framework.

[0949] To analyze questions and generate answers, we use generative AI models provided by OpenAI (e.g., GPT-3).

[0950] Detailed System Description

[0951] User Input

[0952] Users input questions using a dedicated input interface on their smartphone or PC, such as "Where is this wine produced?" or "Please tell me some recipes using this cheese."

[0953] Submit a Question

[0954] The device sends the entered question to a server in the cloud via a specific API endpoint.

[0955] Receiving and parsing questions

[0956] The server receives questions sent by users, analyzes the data, and passes it to a generative AI model (e.g., GPT-3). Specifically, it converts the text data into a format that the AI ​​model can understand.

[0957] Generate answers

[0958] The server uses the analyzed question data to input the generative AI model and generate the optimal answer, such as "This wine is produced in the Bordeaux region of France."

[0959] Returning the answer

[0960] The server returns the generated answer to the user terminal, which displays the returned answer on the screen for the user to confirm.

[0961] Providing and storing feedback

[0962] Users can provide ratings and opinions on generated answers based on their helpfulness and ease of understanding. Feedback is sent to the server and stored in a database. This information is used to continuously learn and improve the generative AI model.

[0963] Specific examples

[0964] For example, if a user types "Where is this wine from?" the system generates the following prompt:

[0965] Prompt Sentence Examples

[0966] Prompt: "Q: Where is this wine from?\nA:"

[0967] The server passes this prompt to the generative AI model, which then generates the answer, "This wine is produced in the Bordeaux region of France." The user receives this information through their device and provides a high rating if they are satisfied with the answer.

[0968] The above is a detailed description of an embodiment of the present invention. The present invention is a system that can improve convenience for customers in a physical store and reduce the burden on employees.

[0969] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0970] Step 1:

[0971] The user inputs a question using the input interface on the terminal. For example, the user can input the question, "Where is this wine produced?" The input data is in text format, and the terminal sends it to the server.

[0972] Step 2:

[0973] The device sends the entered question to a server on the cloud via a specific API endpoint. The sent data is in text format. The server receives this data.

[0974] Step 3:

[0975] The server receives the question sent by the user and analyzes the text data. The analyzed data is converted into a format that can be understood by a generative AI model (e.g., GPT-3). In this step, a prompt sentence is generated. For example, the format is "Prompt: "Q: Where is this wine produced?\nA:"".

[0976] Step 4:

[0977] The server inputs the generated prompt sentence into the generative AI model, which then generates the optimal answer based on the prompt. Data processing involves converting the input text into vector format and applying natural language processing algorithms to generate the answer.

[0978] Step 5:

[0979] The server receives the answer generated by the generative AI model and prepares it to be sent back. The generated answer is in text format. For example, the answer generated might be, "This wine is produced in the Bordeaux region of France."

[0980] Step 6:

[0981] The server sends the generated answer back to the user's terminal, which displays the received answer on the screen, allowing the user to check the answer to the question.

[0982] Step 7:

[0983] Users provide ratings and opinions on the generated answers, such as whether the answer was helpful or easy to understand. The rating data is entered in text format or multiple-choice format.

[0984] Step 8:

[0985] The device sends user feedback to a server, which receives it and stores it in a database. The stored evaluation data is used to train and improve the generative AI model in the future.

[0986] The above is the specific processing flow of the program that realizes this application example. This flow makes it possible to provide quick and accurate answers to customer questions in a physical store.

[0987] 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.

[0988] The present invention is a learning support system in which users (students or teachers) input questions and answers are provided using a generative AI model. Furthermore, the present invention achieves more accurate learning support by combining an emotion engine that recognizes the user's emotions. Specific embodiments of the present invention are described below.

[0989] 1. User inputs a question

[0990] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[0991] 2. Starting Emotion Recognition

[0992] The device recognizes the user's emotions by using facial expressions and voice data when the user inputs a question. This emotion recognition is performed using an emotion engine.

[0993] 3. Sending questions and emotion data

[0994] The device sends the question entered by the user and the accompanying emotion data to the server. The entered data is sent as question text data and emotion data from the emotion engine.

[0995] 4. Receiving and analyzing questions

[0996] The server receives and analyzes questions and emotion data sent from the user's device. The question data is preprocessed and passed to the generative AI model, and the emotion data is used to generate and adjust the answer.

[0997] 5. Generate and refine answers

[0998] The server uses a generative AI model to generate optimal answers to questions. Furthermore, it adjusts the tone and content of the answer based on the user's emotions, based on emotional data from the emotion engine. For example, if the user looks anxious, a more friendly and polite answer will be generated.

[0999] 6. Returning the Response

[1000] The server then sends a tailored response back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Easy to understand, right?"

[1001] 7. View Answers

[1002] The terminal displays the answers returned from the server to the user, who can then check the adjusted answers and deepen their understanding of the learning content.

[1003] 8. Providing Feedback

[1004] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[1005] 9. Sending and Saving Feedback

[1006] The device sends user feedback to the server, which stores it in a database and uses it to subsequently train and improve the generative AI model and emotion engine, thereby improving the accuracy and quality of the entire system.

[1007] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with, "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." Furthermore, if the user looks anxious, the system will include additional explanation such as, "It's okay, let's understand it well." The user will confirm this response and appreciate the detailed and easy-to-understand explanation. In this way, by utilizing the emotion engine, it is possible to provide customized learning support according to the user's emotions.

[1008] In this way, the present invention is a system that allows users to receive prompt and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning. Furthermore, by using an emotion engine, support that takes into consideration the user's emotions can be realized, providing even greater satisfaction and understanding.

[1009] The processing flow will be explained below.

[1010] Program processing flow

[1011] Step 1:

[1012] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[1013] Step 2:

[1014] The device receives the question entered by the user as text data and simultaneously acquires emotion data from the user's facial expressions and voice. The emotion data is used to identify the user's emotion using video and audio analysis approaches.

[1015] Step 3:

[1016] The device transmits the acquired question data and emotion data to the server using a secure communication protocol.

[1017] Step 4:

[1018] The server receives question data and emotion data sent from the user's device. After receiving the data, it performs preprocessing such as tokenizing the text data, and prepares it for passing to the generative AI model and emotion engine.

[1019] Step 5:

[1020] The server uses a generative artificial intelligence model to generate optimal answers based on the question data. The generated answers are based on the underlying learning content.

[1021] Step 6:

[1022] The server uses an emotion engine to tailor the generated responses, specifically adjusting the tone and expression of the responses based on the user's emotion data. For example, if the user expresses anxiety, the responses will be more kind and polite.

[1023] Step 7:

[1024] The server then sends tailored answers back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Was this easy to understand?"

[1025] Step 8:

[1026] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers to deepen their understanding of the learning content.

[1027] Step 9:

[1028] Users provide ratings and opinions (feedback) on the generated answers, such as whether the answers were helpful, easy to understand, or whether they made them feel a certain emotion.

[1029] Step 10:

[1030] The device sends user feedback and emotion data to the server, which will be used to improve the system in the future.

[1031] Step 11:

[1032] The server stores the received feedback and emotion data in a database. This data is used to train and improve the generative AI model and emotion engine. Accumulating data improves the accuracy and quality of the system.

[1033] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, and by using an emotion engine, it provides personalized learning support according to the user's emotions. Furthermore, feedback data can be used to continuously improve the system.

[1034] Example 2

[1035] 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."

[1036] Conventional learning support systems can use artificial intelligence (AI) to provide answers to user questions, but do not generate answers that take the user's emotions into account. This makes it difficult to address the anxiety and confusion users feel while learning, which can reduce the quality and effectiveness of learning. Furthermore, there are also insufficient mechanisms for appropriately accumulating user feedback and reflecting it in subsequent system improvements. The present invention aims to solve these problems.

[1037] 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 means for accepting user input, means for recognizing the user's emotion, means for analyzing the input and emotion data and passing it to a generative AI model, means for generating a response to the user's input using the generative AI model and adjusting the response based on the user's emotion, means for returning the generated response to the user terminal, and means for receiving feedback from the user and using it for training the generative AI model and the emotion recognition means. As a result, by recognizing and responding to the user's emotion, it is possible to provide more appropriate and effective learning support and improve the accuracy and quality of the entire system based on the feedback.

[1038] The "means for accepting user input" refers to a means by which a user inputs questions or information related to learning in the form of text, voice, or the like.

[1039] The "means for recognizing the user's emotions" is a means for analyzing the user's facial expressions and voice and identifying the emotions the user is feeling.

[1040] The "means for analyzing the input and emotion data and passing it to the generative AI model" refers to a means for analyzing the user's question text and voice, as well as the recognized emotion data, and passing it to the generative AI model.

[1041] "Means for using a generative artificial intelligence model to generate a response to a user's input and adjusting the response based on the user's emotions" refers to means for using a generative AI model to generate an optimal response to a user's question and further adjusting the content and tone of the response depending on the user's emotions.

[1042] The "means for returning the generated answer to the user terminal" refers to a means for transmitting the generated and adjusted answer to the user's terminal.

[1043] "Means for receiving feedback from users and using it to train the generative artificial intelligence model and emotion recognition means" refers to means for receiving evaluations and opinions on answers provided by users and using them to improve the performance of the generative artificial intelligence model and emotion recognition engine.

[1044] "Means for storing in a database" means a means for permanently recording user feedback and other relevant data for later analysis and system improvement.

[1045] "Using natural language processing technology" means using technology that applies natural language processing algorithms to the analysis and generation of text data.

[1046] MODE FOR CARRYING OUT THE INVENTION

[1047] The present invention is a system that uses a generative AI model to provide answers to questions about learning entered by a user, and further combines it with an emotion engine to enable customized learning support according to emotions. The following describes in detail the embodiments for implementing this system.

[1048] 1. System Configuration

[1049] This system is broadly composed of a user's device, a server, and a cloud service that includes an emotion engine and a generative AI model.

[1050] User's device

[1051] Users access a dedicated learning support input interface using devices such as PCs or smartphones. This interface is web browser-based software (e.g., implemented using React.js) and is intuitive to operate.

[1052] server

[1053] The server is the central part that receives input from the user and performs the necessary analysis. It is powered by Node.js and Express.js and consists of the following components:

[1054] Input receiving component: Receives question text and sentiment data from the user.

[1055] Preprocessing component: Performs preprocessing such as tokenizing the question text and removing unnecessary symbols.

[1056] Generative AI model invocation component: Invokes a generative AI model, such as OpenAI's GPT-4, to generate answers to questions.

[1057] Emotion adjustment component: Adjusts the generated answers based on the user's emotional data.

[1058] Feedback collection component: receives feedback from users and stores it in a database.

[1059] Emotion Engine

[1060] The emotion engine analyzes the user's facial expressions and voice to identify the emotions they are feeling, using the Emotion API from Microsoft Azure Cognitive Services and the Cloud Speech-to-Text API from Google Cloud.

[1061] Description of software and hardware used

[1062] Webcam and microphone: Hardware that captures a user's facial expressions and records their voice.

[1063] Microsoft Azure Cognitive Services Emotion API: Software for analyzing facial expression data and recognizing emotions.

[1064] Google Cloud Speech-to-Text API: Software for analyzing voice data and recognizing emotions.

[1065] OpenAI GPT-4: A generative AI model that generates answers to questions.

[1066] Node.js and Express.js: Software for performing server-side processing.

[1067] Example of a system

[1068] As a concrete example, if a user types the question "Tell me about the process of photosynthesis," the system will do the following:

[1069] 1. The user types "Please tell me about the process of photosynthesis" into the PC's input interface.

[1070] 2. The device uses a webcam and microphone to capture the user's facial expressions and voice and sends the data to the emotion engine.

[1071] 3. The emotion engine analyzes the captured data and identifies the user's emotion as "anxiety."

[1072] 4. The device sends the entered question and emotion data to the server.

[1073] 5. The server preprocesses the received data and passes it to a generative AI model to generate an answer.

[1074] 6. After the answer is generated, the server takes into account the sentiment data and adds additional encouragement to the answer, such as "It's okay, you'll understand."

[1075] 7. The server sends the tailored response back to the user's terminal.

[1076] 8. The terminal displays the answer for the user to confirm.

[1077] 9. The user rates the answer and provides feedback on whether it was helpful.

[1078] 10. The device sends the feedback to the server, which stores it in a database.

[1079] Prompt Sentence Examples

[1080] Examples of prompts include:

[1081] "Please explain the process of photosynthesis in detail. The user looks anxious."

[1082] In this way, the present invention can provide customized learning assistance that takes into account the user's emotions, improving the quality of learning.

[1083] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1084] Step 1: User enters question

[1085] The user inputs a question related to the study into the input interface of the terminal. The input question text is sent to the interface. For example, by inputting "Please tell me about the process of photosynthesis," the text data is sent to the terminal.

[1086] Step 2: Start Recognizing Emotions

[1087] The device captures the user's facial expressions with a webcam and records their voice with a microphone. The captured data (image data and voice data) is sent to an emotion engine. Specific emotion engines used here include the Emotion API and Cloud Speech-to-Text API. The emotion engine then analyzes the input data and generates emotion data.

[1088] Step 3: Submit your question and sentiment data

[1089] The device sends the entered question text and analyzed emotion data in JSON format to the server. The sent data includes, for example, the following format:

[1090] json

[1091] {

[1092] "question": "Please explain the process of photosynthesis.",

[1093] "emotion": {

[1094] "type": "anxious",

[1095] "confidence": 0.85

[1096] }

[1097] }

[1098] This causes the data to be sent to the server for further processing.

[1099] Step 4: Receive and analyze question and sentiment data

[1100] The server receives the data sent from the device. The received data is divided into question text and emotion data. The server tokenizes the question text and performs preprocessing to remove unnecessary symbols, generating data for analysis. The tokenized question text is then sent to the subsequent generative AI model.

[1101] Step 5: Generate and refine answers

[1102] The server uses a generative AI model to generate answers to questions. Specifically, it invokes OpenAI's GPT-4 model and retrieves the generated answer. For example, it might generate an answer like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen." It then adjusts the tone and content of the answer based on emotional data. If the user shows signs of anxiety, it might add a supporting sentence like, "It's okay, let's understand it."

[1103] Step 6: Submit your response

[1104] The server encodes the adjusted answer in JSON format and returns it to the user's device as an HTTP response, which sends the adjusted answer to the user's device.

[1105] Step 7: View your answers

[1106] The device decodes the response data received from the server and displays it to the user. A React.js component is used for display, allowing the user to view the adjusted response in their browser. For example, the device might display something like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen. It's okay, just make sure you understand it."

[1107] Step 8: Provide feedback

[1108] Users can provide feedback on the displayed answers through a special evaluation form, which includes options such as "helpful" or "confusing."

[1109] Step 9: Submit and save your feedback

[1110] The device encodes the user-provided feedback in JSON format and sends it to the server, which then receives the feedback and stores it in a database. This allows the feedback data to be accumulated and used for future training of the generative AI model and emotion engine.

[1111] (Application example 2)

[1112] 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."

[1113] Conventional learning support systems and customer service systems only provide uniform answers to questions entered by users (students or store clerks), and have the problem of not responding appropriately to the user's emotional state. This can result in insufficient learning benefits or customer service experience, and can lead to a decrease in user satisfaction. Therefore, there is a need for a system that can recognize the user's emotional state in real time and provide appropriate answers based on that.

[1114] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting user input, means for analyzing the input and passing it to a generative AI model, means for generating a response to the user input using the generative AI model, means for returning the generated response to the user terminal, means for receiving feedback from the user and using it for training the generative AI model, means for acquiring user emotion data using emotion recognition means built into the smart device, and means for analyzing the emotion data and adjusting the tone and content of the response. This makes it possible to generate customized responses according to the user's emotional state, thereby improving the quality of learning support and customer service support.

[1115] "User" refers to anyone who uses the system, and specifically includes learners and store clerks who serve customers.

[1116] "Means for accepting input" refers to the interface or device through which a user enters questions or instructions.

[1117] "Means for analyzing and passing to the generative artificial intelligence model" refers to the process for analyzing input data, converting it into an appropriate format, and passing it to the generative artificial intelligence model.

[1118] A "generative artificial intelligence model" refers to an AI model that uses natural language processing technology to generate answers to user input.

[1119] "Means for generating an answer" refers to a function that uses a generative artificial intelligence model to generate the optimal answer to a user's question.

[1120] "Means for returning the generated answer to the user device" refers to the function of sending the answer generated by AI to the device used by the user.

[1121] "Means of receiving feedback and using it for learning" refers to the function of collecting user evaluations and opinions and using them to improve the performance of the generative AI model.

[1122] A "smart device" is a device equipped with communication and computing capabilities, and specifically includes smart glasses, smartphones, and head-mounted displays.

[1123] "Emotion recognition means" refers to technology that uses a camera or microphone installed on the device to recognize emotions from the user's facial expressions and voice.

[1124] "Emotion data" refers to data that indicates the user's emotional state, and includes, for example, emotional states such as joy, anxiety, and anger.

[1125] "Means for adjusting the tone and content of responses" refers to a function that appropriately adjusts the expression and content of generated responses based on emotional data.

[1126] The present invention is embodied as a customer service support system for brick-and-mortar stores. Through an application using smart glasses called "Smart Tutor for Retail," the system provides quick and accurate answers to customer questions. An embodiment of this system will be described in detail below.

[1127] 1. User inputs a question

[1128] The user (store clerk) wears smart glasses and receives questions from customers as voice. For example, if a customer asks, "Please tell me more about this product," the smart glasses' high-performance microphone captures the voice.

[1129] 2. Voice Recognition

[1130] The captured voice data is converted into text by the voice recognition software in the smart glasses, using the Google Cloud Speech-to-Text API.

[1131] 3. Starting Emotion Recognition

[1132] The smart glasses use a built-in camera and microphone to capture the user's facial expressions and voice, thereby collecting emotional data. Emotion recognition is performed using Microsoft Azure Face API and voice analysis software.

[1133] 4. Sending questions and emotion data

[1134] The smart glasses send the converted text data and emotion data to a cloud server, where they are stored via GCP (Google Cloud Platform).

[1135] 5. Parsing Questions and Generating Answers

[1136] The server analyzes the question text using natural language processing (NLP) techniques to generate appropriate answers. The generative AI model used here is OpenAI's GPT-4. The generated answers are adjusted in tone and content based on emotional data.

[1137] 6. Returning and Displaying Tailored Responses

[1138] The server sends the adjusted answer back to the smart glasses, which display the answer on a display for the store clerk to see and also provide an audible response.

[1139] Examples and prompts

[1140] As a concrete example, if a customer asks, "Please tell me more about this product," the process will be as follows:

[1141] Example 1:

[1142] Customer: "Please give me a detailed description of this product."

[1143] Store clerk (through smart glasses): [Camera and microphone capture customer's question]

[1144] System: "This product is made using the latest technology and is particularly durable. For detailed characteristics, please see this URL."

[1145] Also, if the store clerk looks anxious, a more helpful and detailed explanation will be added.

[1146] Example prompt sentence:

[1147] User Question: "Can you give me a detailed description of this product?"

[1148] Prompt to generative AI model: "A customer has asked for a detailed description of a product. Please provide detailed information about the product's features, benefits, and usage, but keep it concise so that the sales associate can easily understand."

[1149] The key feature of this invention is that it recognizes the user's emotions in real time and provides customized responses based on those emotions. This improves the quality of customer service support and increases user satisfaction. In addition, the generative AI model is trained based on the collected feedback, improving the performance of the entire system.

[1150] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1151] Step 1:

[1152] (User inputs question)

[1153] The user (store clerk) wears smart glasses and receives voice questions from customers. For example, the input is voice data such as "Please tell me more about this product."

[1154] Step 2:

[1155] (Voice Recognition)

[1156] The microphone built into the smart glasses captures the received voice data, and then uses voice recognition software (Google Cloud Speech-to-Text API) to convert the voice data into text data. The input is voice data, and the output is text data.

[1157] Step 3:

[1158] (Start of emotion recognition)

[1159] The smart glasses use a built-in camera and microphone to capture the user's facial and voice data, and use the Microsoft Azure Face API to collect the user's emotion data from this data. The input is the captured facial and voice data, and the output is emotion data.

[1160] Step 4:

[1161] (Submitting questions and emotion data)

[1162] The smart glasses send text data and emotion data to a cloud server (GCP). The input is text data and emotion data, and the output is that this data is sent to the cloud server.

[1163] Step 5:

[1164] (Question analysis and answer generation)

[1165] The server analyzes the text data using natural language processing (NLP) techniques. The analyzed question is passed to a generative AI model (e.g., OpenAI's GPT-4) to generate the optimal answer. The tone and content of the generated answer are adjusted based on the emotional data. The input is the text data and emotional data, and the output is the generated answer.

[1166] Step 6:

[1167] (Return and display of adjusted answers)

[1168] The server sends the adjusted answer back to the smart glasses, which display the answer on their display and also speak it aloud. The input is the generated answer, and the output is what the user can see and hear.

[1169] Step 7:

[1170] (Collecting and processing feedback)

[1171] The user (store clerk) inputs feedback on the provided answer through the smart glasses. The smart glasses send this feedback to a cloud server. The server stores the feedback in a database and uses it to train the generative AI model. The input is the feedback data, and the output is the feedback stored in the database and updates to the generative AI model.

[1172] 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.

[1173] 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.

[1174] 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.

[1175] [Fourth embodiment]

[1176] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1177] 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.

[1178] 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).

[1179] 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.

[1180] 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.

[1181] 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).

[1182] 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.

[1183] 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.

[1184] 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.

[1185] 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.

[1186] 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.

[1187] 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.

[1188] 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."

[1189] The present invention is a learning support system in which a user (a student or a teacher) inputs a question and an answer to the question is provided using a generative AI model. Specific embodiments of the present invention are described below.

[1190] 1. User inputs a question

[1191] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[1192] 2. Submit your question

[1193] The device has a function to send questions entered by the user to the server, and the entered data is sent to the server via a specific API endpoint.

[1194] 3. Receiving and analyzing questions

[1195] The server receives questions sent from the user's device, analyzes the data, and passes it to the generative AI model. Specifically, it converts the input text data into a format that the AI ​​model can understand.

[1196] 4. Answer Generation

[1197] The server then passes the analyzed question data to a generative AI model, which processes it to generate the optimal answer to the question. This generative AI model uses natural language processing technology and is capable of generating advanced answers based on past data and learning results.

[1198] 5. Returning the Response

[1199] The server then sends the generated answer back to the user's device, which may include, for example, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[1200] 6. View Answers

[1201] The terminal displays the answer returned from the server to the user, allowing the user to confirm an immediate and appropriate answer to the question.

[1202] 7. Providing Feedback

[1203] Users can provide ratings and opinions on generated answers based on such things as whether the answer was helpful or easy to understand.

[1204] 8. Sending and Saving Feedback

[1205] The device sends user feedback to the server, which stores it in a database and uses it to train and improve the generative AI model, allowing the system to continuously improve its accuracy and quality.

[1206] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." The user can review this answer and provide a high rating if they find it specific and easy to understand. This feedback is stored in the system and used to generate future answers.

[1207] In this way, the present invention is a system that allows users to obtain immediate and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning.

[1208] The processing flow will be explained below.

[1209] Program processing flow

[1210] The specific processing flow of this system will be explained below step by step.

[1211] Step 1:

[1212] Users use a device such as a PC or smartphone to input questions into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[1213] Step 2:

[1214] The device sends the data of the question entered by the user to the server using a dedicated API endpoint, and the entered question is sent in text data format.

[1215] Step 3:

[1216] The server receives questions sent from the user's device. The received question data is preprocessed for analysis and prepared for passing to the generative AI model. Specifically, the text data is tokenized and converted into a format that is easy for the AI ​​model to understand.

[1217] Step 4:

[1218] The server inputs the preprocessed question data into a generative AI model, which generates the optimal answer. This generative AI model uses natural language processing technology to generate answers based on past data and learning results.

[1219] Step 5:

[1220] The server then sends the generated answer back to the user's device, such as, "Photosynthesis is a process primarily performed by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[1221] Step 6:

[1222] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers and deepen their understanding of the learning content.

[1223] Step 7:

[1224] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[1225] Step 8:

[1226] The terminal transmits the feedback from the user to the server, and the feedback data is transmitted in the form of ratings and comments.

[1227] Step 9:

[1228] The server stores the received feedback in a database. This feedback data is used to train and improve the generative AI model. Accumulating feedback data improves the accuracy and quality of the entire system.

[1229] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, maximizing the effectiveness of learning support. It also reduces the burden on teachers and provides learning support tailored to each student.

[1230] Example 1

[1231] 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."

[1232] In modern learning support systems, it is not easy to provide immediate and appropriate answers to questions submitted by users. Furthermore, improving the quality of generated answers and learning models by utilizing subsequent feedback are also important issues. Therefore, a system that can do this efficiently is needed.

[1233] 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.

[1234] In this invention, the server includes means for inputting a question through a user's terminal, means for transmitting the question to the server, means for receiving and analyzing the question at the server and passing it as a prompt to the generative AI model, means for generating an answer to the question using the generative AI model, means for returning the generated answer to the user's terminal, means for displaying the generated answer on the user's terminal, means for receiving feedback from the user, means for transmitting and storing the feedback to the server, and means for using the feedback to train the generative AI model. This allows users to receive instant, high-quality answers to their questions, and the system is continuously improved based on the feedback.

[1235] A "user's terminal" is an electronic device, such as a PC or smartphone, that a user uses to input information.

[1236] A "question" is a textual inquiry that a user enters into the system.

[1237] The "server" is a central processing unit that receives and analyzes instructions sent from the user's device and passes them to the generative AI model.

[1238] A "generative AI model" refers to an artificial intelligence algorithm that uses natural language processing technology to generate optimal answers to input questions.

[1239] A "prompt" is a pre-processed textual instruction that is passed to a generative AI model.

[1240] An "answer" is a text response generated by a generative AI model in response to a user's question.

[1241] "Feedback" refers to the ratings and opinions that users provide on generated answers.

[1242] A "database" is a data storage system for storing and managing data such as feedback.

[1243] "Natural language processing technology" is a computational technology for analyzing, understanding, and generating language that humans use on a daily basis.

[1244] This invention is a learning support system that uses a generative AI model to provide appropriate answers to questions entered by a user. This system consists of multiple components, including a user terminal, a server, and a generative AI model.

[1245] 1. User inputs a question

[1246] Users use a device such as a PC or smartphone to enter questions into a dedicated input interface. Specifically, they open a web browser, access the system's web page, and enter a question into the input form, such as "Please tell me about the process of photosynthesis."

[1247] 2. Submit your question

[1248] The device sends the question entered by the user to the server as an HTTP POST request via an API endpoint. The data is transferred in JSON format, for example:

[1249] json

[1250] {

[1251] "question": "What is the process of photosynthesis?"

[1252] }

[1253] 3. Receiving and analyzing questions

[1254] The server receives the HTTP POST request sent from the device and analyzes the question data contained therein. This analysis includes normalizing the text data and removing unnecessary whitespace and special characters. This converts the text data into a format that the generative AI model can understand.

[1255] 4. Answer Generation

[1256] The server then passes the analyzed question data to a generative AI model as a prompt to generate the optimal answer. This generative AI model uses natural language processing technology, such as OpenAI's GPT-3, which runs on the cloud. The prompt is entered in the following format:

[1257] text

[1258] "Tell me about the process of photosynthesis."

[1259] Generative AI models take into account past data and context to generate answers like:

[1260] text

[1261] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[1262] 5. Returning and Displaying Answers

[1263] The server returns the generated answer in JSON format to the user's device as an HTTP response. The device displays the answer received from the server on the screen and tells the user, for example, as follows:

[1264] text

[1265] "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen."

[1266] 6. Providing and Submitting Feedback

[1267] The user inputs their rating and opinion for the generated answer. For example, they select a rating such as "very helpful" in the rating form. The device sends this rating to the server as an HTTP POST request. An example of the data to be sent is as follows:

[1268] json

[1269] {

[1270] "question": "Explain the process of photosynthesis",

[1271] "answer": "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen.",

[1272] "feedback": "Very helpful"

[1273] }

[1274] 7. Storage and Use of Feedback

[1275] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance and answer accuracy.

[1276] In this way, the system of the present invention allows users to receive instant, high-quality answers to their questions and is continuously improved based on feedback, thereby maximizing the effectiveness of learning support, reducing the burden on teachers, and ultimately improving the quality of learning.

[1277] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1278] Step 1:

[1279] A user inputs a question into a dedicated input interface using a device such as a PC or smartphone. For example, the input question might be text such as "Please tell me about the process of photosynthesis." This input is completed when the user enters the text into an HTML form element and presses the submit button.

[1280] Input: User-provided question text

[1281] Output: Question text in terminal input form

[1282] Step 2:

[1283] The device sends the question entered by the user to the server by sending an HTTP POST request to the API endpoint. The question text is encoded in JSON format and sent.

[1284] Input: User question text

[1285] Output: JSON formatted data sent to the server

[1286] Step 3:

[1287] The server receives an HTTP POST request sent from the device, which contains the user's question. At this stage, the server extracts the question text from the request and normalizes it, removing unnecessary whitespace and special characters.

[1288] Input: JSON format data received from the terminal

[1289] Output: Normalized question text

[1290] Step 4:

[1291] The server passes the normalized question text as a prompt to a generative AI model, which uses natural language processing techniques, for example. The server sends the prompt to the generative AI model, which then generates an answer.

[1292] Input: Normalized question text

[1293] Output: Generated answer text

[1294] Step 5:

[1295] The server packages the answer received from the generative AI model in JSON format and returns it to the user's device as an HTTP response, which includes the generated answer text.

[1296] Input: Answer text from the generative AI model

[1297] Output: JSON formatted response data

[1298] Step 6:

[1299] The terminal processes the HTTP response received from the server and displays the generated answer to the user in HTML format. Using JavaScript, the answer text is inserted into the specified DOM element on the browser.

[1300] Input: JSON format response data from the server

[1301] Output: Answer text displayed in the browser

[1302] Step 7:

[1303] The user can enter feedback on the displayed answer by using the evaluation form to select a rating such as "helpful" or "easy to understand" and enter comments if necessary.

[1304] Input: User feedback on the generated answer

[1305] Output: Feedback text

[1306] Step 8:

[1307] The device sends the user-entered feedback to the server, again using an HTTP POST request, with the feedback encoded in JSON format.

[1308] Input: User feedback text

[1309] Output: Feedback data sent to the server in JSON format.

[1310] Step 9:

[1311] The server stores the received feedback data in a database, which is used to further train the generative AI model and improve the system's performance.

[1312] Input: Feedback data received from the device in JSON format

[1313] Output: Feedback data stored in a database

[1314] (Application example 1)

[1315] 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."

[1316] In traditional brick-and-mortar stores, when customers wanted detailed information about a product or a guided tour of the store, it was difficult to provide them with the information quickly and accurately. Furthermore, because the store required the assistance of an employee, responses were often delayed during busy periods, resulting in lower customer satisfaction. Furthermore, the quality and speed of responses to customer questions depended on the experience and knowledge of the employee, which meant that consistent service could be inconsistent. To solve these issues, a system was needed that would allow customers to input questions and receive instant answers.

[1317] 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.

[1318] In this invention, the server includes a means for accepting user input, a means for analyzing the input and passing it to the generative AI model, and a means for returning the generated answer to the user terminal, thereby enabling quick and accurate answers to questions about product explanations, store guides, and recipe suggestions in a physical store to be provided.

[1319] The "means for accepting user input" is a function that allows a user to input questions or requests to the terminal via a dedicated input interface.

[1320] The "means for analyzing input and passing it to the generative artificial intelligence model" is a processing function for analyzing data input by a user, converting it into a format that can be understood by the artificial intelligence model, and passing it on.

[1321] "Means for generating answers to user input using a generative artificial intelligence model" refers to algorithms and techniques that allow the artificial intelligence model to generate optimal answers based on analyzed input data.

[1322] "Means for returning the generated answer to the user terminal" refers to communication and display technology for transferring and displaying the answer generated by the artificial intelligence model to the user's terminal.

[1323] "Means for receiving feedback from users and using it to train the generative AI model" refers to data collection and learning technology that collects evaluations and opinions on answers provided by users and uses them to improve and enhance the accuracy of the AI ​​model.

[1324] "Means of providing answers to questions regarding product descriptions, store guidance, and recipe suggestions within a physical store" refers to functions and technologies that allow customers to ask questions regarding product and store information and recipe suggestions in a physical store, and for an AI model to provide immediate and appropriate answers.

[1325] "Means for storing data in a database" refers to the storage systems and technologies used to organize and securely store collected data and feedback.

[1326] "Using natural language processing technology" refers to the process of analyzing text entered by a user and generating an answer using algorithms and technologies for understanding and generating human language.

[1327] A specific embodiment of the present invention will be described in detail below. This invention is a system that allows users (customers) to ask questions about products, store information, and recipe suggestions in a physical store and receive automatic answers.

[1328] System Configuration

[1329] Hardware

[1330] The server uses a virtual machine on the cloud (e.g., AWS EC2, Google Cloud VM), and the user's device is a smartphone (iOS / Android) or PC.

[1331] software

[1332] The programming language used is Python.

[1333] Use Flask as a web framework.

[1334] To analyze questions and generate answers, we use generative AI models provided by OpenAI (e.g., GPT-3).

[1335] Detailed System Description

[1336] User Input

[1337] Users input questions using a dedicated input interface on their smartphone or PC, such as "Where is this wine produced?" or "Please tell me some recipes using this cheese."

[1338] Submit a Question

[1339] The device sends the entered question to a server in the cloud via a specific API endpoint.

[1340] Receiving and parsing questions

[1341] The server receives questions sent by users, analyzes the data, and passes it to a generative AI model (e.g., GPT-3). Specifically, it converts the text data into a format that the AI ​​model can understand.

[1342] Generate answers

[1343] The server uses the analyzed question data to input the generative AI model and generate the optimal answer, such as "This wine is produced in the Bordeaux region of France."

[1344] Returning the answer

[1345] The server returns the generated answer to the user terminal, which displays the returned answer on the screen for the user to confirm.

[1346] Providing and storing feedback

[1347] Users can provide ratings and opinions on generated answers based on their helpfulness and ease of understanding. Feedback is sent to the server and stored in a database. This information is used to continuously learn and improve the generative AI model.

[1348] Specific examples

[1349] For example, if a user types "Where is this wine from?" the system generates the following prompt:

[1350] Prompt Sentence Examples

[1351] Prompt: "Q: Where is this wine from?\nA:"

[1352] The server passes this prompt to the generative AI model, which then generates the answer, "This wine is produced in the Bordeaux region of France." The user receives this information through their device and provides a high rating if they are satisfied with the answer.

[1353] The above is a detailed description of an embodiment of the present invention. The present invention is a system that can improve convenience for customers in a physical store and reduce the burden on employees.

[1354] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1355] Step 1:

[1356] The user inputs a question using the input interface on the terminal. For example, the user can input the question, "Where is this wine produced?" The input data is in text format, and the terminal sends it to the server.

[1357] Step 2:

[1358] The device sends the entered question to a server on the cloud via a specific API endpoint. The sent data is in text format. The server receives this data.

[1359] Step 3:

[1360] The server receives the question sent by the user and analyzes the text data. The analyzed data is converted into a format that can be understood by a generative AI model (e.g., GPT-3). In this step, a prompt sentence is generated. For example, the format is "Prompt: "Q: Where is this wine produced?\nA:"".

[1361] Step 4:

[1362] The server inputs the generated prompt sentence into the generative AI model, which then generates the optimal answer based on the prompt. Data processing involves converting the input text into vector format and applying natural language processing algorithms to generate the answer.

[1363] Step 5:

[1364] The server receives the answer generated by the generative AI model and prepares it to be sent back. The generated answer is in text format. For example, the answer generated might be, "This wine is produced in the Bordeaux region of France."

[1365] Step 6:

[1366] The server sends the generated answer back to the user's terminal, which displays the received answer on the screen, allowing the user to check the answer to the question.

[1367] Step 7:

[1368] Users provide ratings and opinions on the generated answers, such as whether the answer was helpful or easy to understand. The rating data is entered in text format or multiple-choice format.

[1369] Step 8:

[1370] The device sends user feedback to a server, which receives it and stores it in a database. The stored evaluation data is used to train and improve the generative AI model in the future.

[1371] The above is the specific processing flow of the program that realizes this application example. This flow makes it possible to provide quick and accurate answers to customer questions in a physical store.

[1372] 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.

[1373] The present invention is a learning support system in which users (students or teachers) input questions and answers are provided using a generative AI model. Furthermore, the present invention achieves more accurate learning support by combining an emotion engine that recognizes the user's emotions. Specific embodiments of the present invention are described below.

[1374] 1. User inputs a question

[1375] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[1376] 2. Starting Emotion Recognition

[1377] The device recognizes the user's emotions by using facial expressions and voice data when the user inputs a question. This emotion recognition is performed using an emotion engine.

[1378] 3. Sending questions and emotion data

[1379] The device sends the question entered by the user and the accompanying emotion data to the server. The entered data is sent as question text data and emotion data from the emotion engine.

[1380] 4. Receiving and analyzing questions

[1381] The server receives and analyzes questions and emotion data sent from the user's device. The question data is preprocessed and passed to the generative AI model, and the emotion data is used to generate and adjust the answer.

[1382] 5. Generate and refine answers

[1383] The server uses a generative AI model to generate optimal answers to questions. Furthermore, it adjusts the tone and content of the answer based on the user's emotions, based on emotional data from the emotion engine. For example, if the user looks anxious, a more friendly and polite answer will be generated.

[1384] 6. Returning the Response

[1385] The server then sends a tailored response back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Easy to understand, right?"

[1386] 7. View Answers

[1387] The terminal displays the answers returned from the server to the user, who can then check the adjusted answers and deepen their understanding of the learning content.

[1388] 8. Providing Feedback

[1389] The user provides feedback and ratings for the generated answers, based on criteria such as whether the answer was helpful or easy to understand.

[1390] 9. Sending and Saving Feedback

[1391] The device sends user feedback to the server, which stores it in a database and uses it to subsequently train and improve the generative AI model and emotion engine, thereby improving the accuracy and quality of the entire system.

[1392] For example, if a user inputs a question such as "Please tell me about the process of photosynthesis," the system will respond with, "Photosynthesis is a process mainly carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen." Furthermore, if the user looks anxious, the system will include additional explanation such as, "It's okay, let's understand it well." The user will confirm this response and appreciate the detailed and easy-to-understand explanation. In this way, by utilizing the emotion engine, it is possible to provide customized learning support according to the user's emotions.

[1393] In this way, the present invention is a system that allows users to receive prompt and appropriate answers to their questions, maximizing the effectiveness of learning support, reducing the burden on teachers, and improving the quality of learning. Furthermore, by using an emotion engine, support that takes into consideration the user's emotions can be realized, providing even greater satisfaction and understanding.

[1394] The processing flow will be explained below.

[1395] Program processing flow

[1396] Step 1:

[1397] Users use a device such as a PC or smartphone to input questions related to their studies into a dedicated input interface, such as "Please tell me about the process of photosynthesis."

[1398] Step 2:

[1399] The device receives the question entered by the user as text data and simultaneously acquires emotion data from the user's facial expressions and voice. The emotion data is used to identify the user's emotion using video and audio analysis approaches.

[1400] Step 3:

[1401] The device transmits the acquired question data and emotion data to the server using a secure communication protocol.

[1402] Step 4:

[1403] The server receives question data and emotion data sent from the user's device. After receiving the data, it performs preprocessing such as tokenizing the text data, and prepares it for passing to the generative AI model and emotion engine.

[1404] Step 5:

[1405] The server uses a generative artificial intelligence model to generate optimal answers based on the question data. The generated answers are based on the underlying learning content.

[1406] Step 6:

[1407] The server uses an emotion engine to tailor the generated responses, specifically adjusting the tone and expression of the responses based on the user's emotion data. For example, if the user expresses anxiety, the responses will be more kind and polite.

[1408] Step 7:

[1409] The server then sends tailored answers back to the user's device, such as, "Photosynthesis is a process primarily carried out by plants, which uses sunlight, carbon dioxide, and water to produce glucose and oxygen. Was this easy to understand?"

[1410] Step 8:

[1411] The terminal receives the answers sent back from the server and displays them to the user, who can then check the answers to deepen their understanding of the learning content.

[1412] Step 9:

[1413] Users provide ratings and opinions (feedback) on the generated answers, such as whether the answers were helpful, easy to understand, or whether they made them feel a certain emotion.

[1414] Step 10:

[1415] The device sends user feedback and emotion data to the server, which will be used to improve the system in the future.

[1416] Step 11:

[1417] The server stores the received feedback and emotion data in a database. This data is used to train and improve the generative AI model and emotion engine. Accumulating data improves the accuracy and quality of the system.

[1418] Through these processing steps, the system enables users to get quick and appropriate answers to their questions, and by using an emotion engine, it provides personalized learning support according to the user's emotions. Furthermore, feedback data can be used to continuously improve the system.

[1419] Example 2

[1420] 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."

[1421] Conventional learning support systems can use artificial intelligence (AI) to provide answers to user questions, but do not generate answers that take the user's emotions into account. This makes it difficult to address the anxiety and confusion users feel while learning, which can reduce the quality and effectiveness of learning. Furthermore, there are also insufficient mechanisms for appropriately accumulating user feedback and reflecting it in subsequent system improvements. The present invention aims to solve these problems.

[1422] 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 means for accepting user input, means for recognizing the user's emotion, means for analyzing the input and emotion data and passing it to a generative AI model, means for generating a response to the user's input using the generative AI model and adjusting the response based on the user's emotion, means for returning the generated response to the user terminal, and means for receiving feedback from the user and using it for training the generative AI model and the emotion recognition means. As a result, by recognizing and responding to the user's emotion, it is possible to provide more appropriate and effective learning support and improve the accuracy and quality of the entire system based on the feedback.

[1423] The "means for accepting user input" refers to a means by which a user inputs questions or information related to learning in the form of text, voice, or the like.

[1424] The "means for recognizing the user's emotions" is a means for analyzing the user's facial expressions and voice and identifying the emotions the user is feeling.

[1425] The "means for analyzing the input and emotion data and passing it to the generative AI model" refers to a means for analyzing the user's question text and voice, as well as the recognized emotion data, and passing it to the generative AI model.

[1426] "Means for using a generative artificial intelligence model to generate a response to a user's input and adjusting the response based on the user's emotions" refers to means for using a generative AI model to generate an optimal response to a user's question and further adjusting the content and tone of the response depending on the user's emotions.

[1427] The "means for returning the generated answer to the user terminal" refers to a means for transmitting the generated and adjusted answer to the user's terminal.

[1428] "Means for receiving feedback from users and using it to train the generative artificial intelligence model and emotion recognition means" refers to means for receiving evaluations and opinions on answers provided by users and using them to improve the performance of the generative artificial intelligence model and emotion recognition engine.

[1429] "Means for storing in a database" means a means for permanently recording user feedback and other relevant data for later analysis and system improvement.

[1430] "Using natural language processing technology" means using technology that applies natural language processing algorithms to the analysis and generation of text data.

[1431] MODE FOR CARRYING OUT THE INVENTION

[1432] The present invention is a system that uses a generative AI model to provide answers to questions about learning entered by a user, and further combines it with an emotion engine to enable customized learning support according to emotions. The following describes in detail the embodiments for implementing this system.

[1433] 1. System Configuration

[1434] This system is broadly composed of a user's device, a server, and a cloud service that includes an emotion engine and a generative AI model.

[1435] User's device

[1436] Users access a dedicated learning support input interface using devices such as PCs or smartphones. This interface is web browser-based software (e.g., implemented using React.js) and is intuitive to operate.

[1437] server

[1438] The server is the central part that receives input from the user and performs the necessary analysis. It is powered by Node.js and Express.js and consists of the following components:

[1439] Input receiving component: Receives question text and sentiment data from the user.

[1440] Preprocessing component: Performs preprocessing such as tokenizing the question text and removing unnecessary symbols.

[1441] Generative AI model invocation component: Invokes a generative AI model, such as OpenAI's GPT-4, to generate answers to questions.

[1442] Emotion adjustment component: Adjusts the generated answers based on the user's emotional data.

[1443] Feedback collection component: receives feedback from users and stores it in a database.

[1444] Emotion Engine

[1445] The emotion engine analyzes the user's facial expressions and voice to identify the emotions they are feeling, using the Emotion API from Microsoft Azure Cognitive Services and the Cloud Speech-to-Text API from Google Cloud.

[1446] Description of software and hardware used

[1447] Webcam and microphone: Hardware that captures a user's facial expressions and records their voice.

[1448] Microsoft Azure Cognitive Services Emotion API: Software for analyzing facial expression data and recognizing emotions.

[1449] Google Cloud Speech-to-Text API: Software for analyzing voice data and recognizing emotions.

[1450] OpenAI GPT-4: A generative AI model that generates answers to questions.

[1451] Node.js and Express.js: Software for performing server-side processing.

[1452] Example of a system

[1453] As a concrete example, if a user types the question "Tell me about the process of photosynthesis," the system will do the following:

[1454] 1. The user types "Please tell me about the process of photosynthesis" into the PC's input interface.

[1455] 2. The device uses a webcam and microphone to capture the user's facial expressions and voice and sends the data to the emotion engine.

[1456] 3. The emotion engine analyzes the captured data and identifies the user's emotion as "anxiety."

[1457] 4. The device sends the entered question and emotion data to the server.

[1458] 5. The server preprocesses the received data and passes it to a generative AI model to generate an answer.

[1459] 6. After the answer is generated, the server takes into account the sentiment data and adds additional encouragement to the answer, such as "It's okay, you'll understand."

[1460] 7. The server sends the tailored response back to the user's terminal.

[1461] 8. The terminal displays the answer for the user to confirm.

[1462] 9. The user rates the answer and provides feedback on whether it was helpful.

[1463] 10. The device sends the feedback to the server, which stores it in a database.

[1464] Prompt Sentence Examples

[1465] Examples of prompts include:

[1466] "Please explain the process of photosynthesis in detail. The user looks anxious."

[1467] In this way, the present invention can provide customized learning assistance that takes into account the user's emotions, improving the quality of learning.

[1468] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1469] Step 1: User enters question

[1470] The user inputs a question related to the study into the input interface of the terminal. The input question text is sent to the interface. For example, by inputting "Please tell me about the process of photosynthesis," the text data is sent to the terminal.

[1471] Step 2: Start Recognizing Emotions

[1472] The device captures the user's facial expressions with a webcam and records their voice with a microphone. The captured data (image data and voice data) is sent to an emotion engine. Specific emotion engines used here include the Emotion API and Cloud Speech-to-Text API. The emotion engine then analyzes the input data and generates emotion data.

[1473] Step 3: Submit your question and sentiment data

[1474] The device sends the entered question text and analyzed emotion data in JSON format to the server. The sent data includes, for example, the following format:

[1475] json

[1476] {

[1477] "question": "Please explain the process of photosynthesis.",

[1478] "emotion": {

[1479] "type": "anxious",

[1480] "confidence": 0.85

[1481] }

[1482] }

[1483] This causes the data to be sent to the server for further processing.

[1484] Step 4: Receive and analyze question and sentiment data

[1485] The server receives the data sent from the device. The received data is divided into question text and emotion data. The server tokenizes the question text and performs preprocessing to remove unnecessary symbols, generating data for analysis. The tokenized question text is then sent to the subsequent generative AI model.

[1486] Step 5: Generate and refine answers

[1487] The server uses a generative AI model to generate answers to questions. Specifically, it invokes OpenAI's GPT-4 model and retrieves the generated answer. For example, it might generate an answer like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen." It then adjusts the tone and content of the answer based on emotional data. If the user shows signs of anxiety, it might add a supporting sentence like, "It's okay, let's understand it."

[1488] Step 6: Submit your response

[1489] The server encodes the adjusted answer in JSON format and returns it to the user's device as an HTTP response, which sends the adjusted answer to the user's device.

[1490] Step 7: View your answers

[1491] The device decodes the response data received from the server and displays it to the user. A React.js component is used for display, allowing the user to view the adjusted response in their browser. For example, the device might display something like, "Photosynthesis is a process primarily performed by plants, using sunlight, carbon dioxide, and water to produce glucose and oxygen. It's okay, just make sure you understand it."

[1492] Step 8: Provide feedback

[1493] Users can provide feedback on the displayed answers through a special evaluation form, which includes options such as "helpful" or "confusing."

[1494] Step 9: Submit and save your feedback

[1495] The device encodes the user-provided feedback in JSON format and sends it to the server, which then receives the feedback and stores it in a database. This allows the feedback data to be accumulated and used for future training of the generative AI model and emotion engine.

[1496] (Application example 2)

[1497] 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."

[1498] Conventional learning support systems and customer service systems only provide uniform answers to questions entered by users (students or store clerks), and have the problem of not responding appropriately to the user's emotional state. This can result in insufficient learning benefits or customer service experience, and can lead to a decrease in user satisfaction. Therefore, there is a need for a system that can recognize the user's emotional state in real time and provide appropriate answers based on that.

[1499] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for accepting user input, means for analyzing the input and passing it to a generative AI model, means for generating a response to the user input using the generative AI model, means for returning the generated response to the user terminal, means for receiving feedback from the user and using it for training the generative AI model, means for acquiring user emotion data using emotion recognition means built into the smart device, and means for analyzing the emotion data and adjusting the tone and content of the response. This makes it possible to generate customized responses according to the user's emotional state, thereby improving the quality of learning support and customer service support.

[1500] "User" refers to anyone who uses the system, and specifically includes learners and store clerks who serve customers.

[1501] "Means for accepting input" refers to the interface or device through which a user enters questions or instructions.

[1502] "Means for analyzing and passing to the generative artificial intelligence model" refers to the process for analyzing input data, converting it into an appropriate format, and passing it to the generative artificial intelligence model.

[1503] A "generative artificial intelligence model" refers to an AI model that uses natural language processing technology to generate answers to user input.

[1504] "Means for generating an answer" refers to a function that uses a generative artificial intelligence model to generate the optimal answer to a user's question.

[1505] "Means for returning the generated answer to the user device" refers to the function of sending the answer generated by AI to the device used by the user.

[1506] "Means of receiving feedback and using it for learning" refers to the function of collecting user evaluations and opinions and using them to improve the performance of the generative AI model.

[1507] A "smart device" is a device equipped with communication and computing capabilities, and specifically includes smart glasses, smartphones, and head-mounted displays.

[1508] "Emotion recognition means" refers to technology that uses a camera or microphone installed on the device to recognize emotions from the user's facial expressions and voice.

[1509] "Emotion data" refers to data that indicates the user's emotional state, and includes, for example, emotional states such as joy, anxiety, and anger.

[1510] "Means for adjusting the tone and content of responses" refers to a function that appropriately adjusts the expression and content of generated responses based on emotional data.

[1511] The present invention is embodied as a customer service support system for brick-and-mortar stores. Through an application using smart glasses called "Smart Tutor for Retail," the system provides quick and accurate answers to customer questions. An embodiment of this system will be described in detail below.

[1512] 1. User inputs a question

[1513] The user (store clerk) wears smart glasses and receives questions from customers as voice. For example, if a customer asks, "Please tell me more about this product," the smart glasses' high-performance microphone captures the voice.

[1514] 2. Voice Recognition

[1515] The captured voice data is converted into text by the voice recognition software in the smart glasses, using the Google Cloud Speech-to-Text API.

[1516] 3. Starting Emotion Recognition

[1517] The smart glasses use a built-in camera and microphone to capture the user's facial expressions and voice, thereby collecting emotional data. Emotion recognition is performed using Microsoft Azure Face API and voice analysis software.

[1518] 4. Sending questions and emotion data

[1519] The smart glasses send the converted text data and emotion data to a cloud server, where they are stored via GCP (Google Cloud Platform).

[1520] 5. Parsing Questions and Generating Answers

[1521] The server analyzes the question text using natural language processing (NLP) techniques to generate appropriate answers. The generative AI model used here is OpenAI's GPT-4. The generated answers are adjusted in tone and content based on emotional data.

[1522] 6. Returning and Displaying Tailored Responses

[1523] The server sends the adjusted answer back to the smart glasses, which display the answer on a display for the store clerk to see and also provide an audible response.

[1524] Examples and prompts

[1525] As a concrete example, if a customer asks, "Please tell me more about this product," the process will be as follows:

[1526] Example 1:

[1527] Customer: "Please give me a detailed description of this product."

[1528] Store clerk (through smart glasses): [Camera and microphone capture customer's question]

[1529] System: "This product is made using the latest technology and is particularly durable. For detailed characteristics, please see this URL."

[1530] Also, if the store clerk looks anxious, a more helpful and detailed explanation will be added.

[1531] Example prompt sentence:

[1532] User Question: "Can you give me a detailed description of this product?"

[1533] Prompt to generative AI model: "A customer has asked for a detailed description of a product. Please provide detailed information about the product's features, benefits, and usage, but keep it concise so that the sales associate can easily understand."

[1534] The key feature of this invention is that it recognizes the user's emotions in real time and provides customized responses based on those emotions. This improves the quality of customer service support and increases user satisfaction. In addition, the generative AI model is trained based on the collected feedback, improving the performance of the entire system.

[1535] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1536] Step 1:

[1537] (User inputs question)

[1538] The user (store clerk) wears smart glasses and receives voice questions from customers. For example, the input is voice data such as "Please tell me more about this product."

[1539] Step 2:

[1540] (Voice Recognition)

[1541] The microphone built into the smart glasses captures the received voice data, and then uses voice recognition software (Google Cloud Speech-to-Text API) to convert the voice data into text data. The input is voice data, and the output is text data.

[1542] Step 3:

[1543] (Start of emotion recognition)

[1544] The smart glasses use a built-in camera and microphone to capture the user's facial and voice data, and use the Microsoft Azure Face API to collect the user's emotion data from this data. The input is the captured facial and voice data, and the output is emotion data.

[1545] Step 4:

[1546] (Submitting questions and emotion data)

[1547] The smart glasses send text data and emotion data to a cloud server (GCP). The input is text data and emotion data, and the output is that this data is sent to the cloud server.

[1548] Step 5:

[1549] (Question analysis and answer generation)

[1550] The server analyzes the text data using natural language processing (NLP) techniques. The analyzed question is passed to a generative AI model (e.g., OpenAI's GPT-4) to generate the optimal answer. The tone and content of the generated answer are adjusted based on the emotional data. The input is the text data and emotional data, and the output is the generated answer.

[1551] Step 6:

[1552] (Return and display of adjusted answers)

[1553] The server sends the adjusted answer back to the smart glasses, which display the answer on their display and also speak it aloud. The input is the generated answer, and the output is what the user can see and hear.

[1554] Step 7:

[1555] (Collecting and processing feedback)

[1556] The user (store clerk) inputs feedback on the provided answer through the smart glasses. The smart glasses send this feedback to a cloud server. The server stores the feedback in a database and uses it to train the generative AI model. The input is the feedback data, and the output is the feedback stored in the database and updates to the generative AI model.

[1557] 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.

[1558] 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.

[1559] 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.

[1560] 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.

[1561] 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.

[1562] 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.

[1563] 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).

[1564] 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.

[1565] 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."

[1566] 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.

[1567] 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).

[1568] 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.

[1569] 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.

[1570] 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.

[1571] 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.

[1572] 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.

[1573] 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.

[1574] 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.

[1575] 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.

[1576] 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.

[1577] 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.

[1578] The following is further disclosed regarding the above embodiment.

[1579] (Claim 1)

[1580] means for accepting user input;

[1581] A means for analyzing the input and passing it to a generative artificial intelligence model;

[1582] means for generating an answer to a user's input using the generative artificial intelligence model;

[1583] means for returning the generated answer to the user terminal;

[1584] means for receiving feedback from a user and utilizing it in training the generative artificial intelligence model;

[1585] A system including:

[1586] (Claim 2)

[1587] 10. The system of claim 1, further comprising means for storing said feedback in a database.

[1588] (Claim 3)

[1589] The system of claim 1 , wherein the generative artificial intelligence model uses natural language processing techniques.

[1590] "Example 1"

[1591] (Claim 1)

[1592] means for inputting a question via a user terminal;

[1593] means for sending a query to a server;

[1594] A means to receive and analyze questions on the server and pass them as prompts to the generative AI model;

[1595] a means for generating answers to questions using a generative AI model;

[1596] means for returning the generated answer to the user's terminal;

[1597] means for displaying the generated answer on the user's device;

[1598] a means for receiving feedback from users;

[1599] means for transmitting and storing the feedback on a server;

[1600] means for utilizing the feedback to train a generative AI model;

[1601] A system including:

[1602] (Claim 2)

[1603] 10. The system of claim 1, further comprising means for storing said feedback in a database.

[1604] (Claim 3)

[1605] The system of claim 1 , wherein the generative AI model uses natural language processing techniques.

[1606] "Application Example 1"

[1607] (Claim 1)

[1608] means for accepting user input;

[1609] A means for analyzing the input and passing it to a generative artificial intelligence model;

[1610] means for generating an answer to a user's input using the generative artificial intelligence model;

[1611] means for returning the generated answer to the user terminal;

[1612] means for receiving feedback from a user and utilizing it in training the generative artificial intelligence model;

[1613] A means to provide answers to questions in order to provide product descriptions, store information, and recipe suggestions in physical stores,

[1614] A system including:

[1615] (Claim 2)

[1616] 10. The system of claim 1, further comprising means for storing said feedback in a database.

[1617] (Claim 3)

[1618] The system of claim 1 , wherein the generative artificial intelligence model uses natural language processing techniques.

[1619] "Example 2: Combining Emotion Engines"

[1620] (Claim 1)

[1621] means for accepting user input;

[1622] means for recognizing a user's emotion;

[1623] A means for analyzing the input and emotion data and passing it to a generative artificial intelligence model;

[1624] means for generating a response to a user's input using the generative artificial intelligence model and adjusting the response based on the user's emotions;

[1625] means for returning the generated answer to the user terminal;

[1626] means for receiving feedback from a user and using it to train the generative artificial intelligence model and the emotion recognition means;

[1627] A system including:

[1628] (Claim 2)

[1629] 10. The system of claim 1, further comprising means for storing said feedback in a database.

[1630] (Claim 3)

[1631] The system of claim 1 , wherein the generative artificial intelligence model uses natural language processing techniques.

[1632] "Application example 2 when combining emotion engines"

[1633] (Claim 1)

[1634] means for accepting user input;

[1635] A means for analyzing the input and passing it to a generative artificial intelligence model;

[1636] means for generating an answer to a user's input using the generative artificial intelligence model;

[1637] means for returning the generated answer to the user terminal;

[1638] means for receiving feedback from a user and utilizing it in training the generative artificial intelligence model;

[1639] A means for acquiring emotion data of a user using an emotion recognition means built into the smart device;

[1640] means for analyzing said emotional data to adjust the tone and content of responses;

[1641] A system including:

[1642] (Claim 2)

[1643] 10. The system of claim 1, further comprising means for storing said feedback in a database.

[1644] (Claim 3)

[1645] The system of claim 1 , wherein the generative artificial intelligence model uses natural language processing techniques. [Explanation of symbols]

[1646] 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. means for accepting user input; A means for analyzing the input and passing it to a generative artificial intelligence model; means for generating an answer to a user's input using the generative artificial intelligence model; means for returning the generated answer to the user terminal; means for receiving feedback from a user and utilizing it in training the generative artificial intelligence model; A system including:

2. The system of claim 1 further comprising means for storing said feedback in a database.

3. The system of claim 1 , wherein the generative artificial intelligence model uses natural language processing techniques.

Images