system

Abstract

Provide a system. The method includes: generating a response to a user query using an AI model; means for recording received user queries and user IDs; a means for generating a response to the received query and returning the response in JSON format; A system including:

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 educational institutions, it is extremely important to respond quickly and accurately to inquiries from faculty, students, and other stakeholders. However, there are limitations on human resources, and delays in responding to inquiries can occur, especially when it is difficult to respond at night or on holidays. This increases the burden on staff and leads to a decrease in customer satisfaction, which is an issue. [Means for solving the problem]

[0005] The present invention solves these problems by providing a chatbot system using an AI model. Specifically, the system includes a means for automatically generating responses to user queries and recording the received queries and user IDs. It also includes a response generation means that self-learns using past inquiry history, and operates in a server environment that can operate 24 hours a day, reducing the workload on educational institution staff and improving customer satisfaction.

[0006] "User" refers to a person or entity that submits queries to the system and receives responses.

[0007] A "query" refers to a question or request for information that a user poses to a system.

[0008] "Response" refers to information that a system generates in response to a user query.

[0009] "AI Model" refers to a computational model that uses artificial intelligence techniques to generate appropriate responses to user queries.

[0010] "Recording" refers to the act of saving information such as received user queries and user IDs as a log.

[0011] "JSON format" is an abbreviation for JavaScript (registered trademark) Object Notation, and refers to a lightweight data exchange format.

[0012] "Server" refers to a computer system for handling user interactions over a network.

[0013] "Self-learning" refers to the process of using past data and query history to improve system performance and response accuracy.

[0014] "Query history" refers to a record of queries received from users in the past and the responses to those queries.

[0015] "24-hour operation" means that the system has the ability to operate continuously at all times, day or night. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] The present invention is a system that uses AI models to automate customer support for inquiries from faculty, students, and other stakeholders at educational institutions. The system includes the following main functions:

[0038] 1. Receiving queries from users

[0039] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0040] 2. Query Parsing and Logging

[0041] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0042] 3. Generating a Response

[0043] The server generates appropriate responses to user queries using an AI model that is self-learning based on past query history and is constantly tuned to provide the best possible response.

[0044] 4. Returning a Response

[0045] The generated response is converted to JSON format and sent to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0046] Natural language explanation of the process

[0047] Query reception

[0048] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[0049] json

[0050] {

[0051] "user_id": "student123",

[0052] "query": "When is the next math test?"

[0053] }

[0054] Query Parsing and Logging

[0055] The server receives this request and parses the user_id and query values, which are logged as follows:

[0056] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0057] Generating a response

[0058] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model is actually self-trained based on past query data, which improves the accuracy of the answer given to the user. An example response might be the following text:

[0059] The next math test is next Friday.

[0060] Returning a response

[0061] The server converts the generated response into JSON format and returns it to the user's device as an HTTP response, which the user sees on their device and sees something like this:

[0062] json

[0063] {

[0064] "response": "The next math test is next Friday."

[0065] }

[0066] The system operates in a 24-hour server environment, allowing users to make inquiries at any time, reducing the burden on educational institution staff and providing quick and accurate responses.

[0067] The processing flow will be explained below.

[0068] Step 1:

[0069] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[0070] json

[0071] {

[0072] "user_id": "student123",

[0073] "query": "When is the next math test?"

[0074] }

[0075] Step 2:

[0076] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[0077] Step 3:

[0078] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[0079] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0080] Step 4:

[0081] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. This function uses a self-trained AI model based on past query history to return the best possible response. In this simple example, the query is returned as the response.

[0082] Step 5:

[0083] The server converts the generated response into JSON format by using the jsonify function to package the generated text as a JSON object, for example generating a JSON response like this:

[0084] json

[0085] {

[0086] "response": "The next math test is next Friday."

[0087] }

[0088] Step 6:

[0089] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[0090] AI chatbot response: The next math test is next Friday.

[0091] The above processing steps enable the institution to provide quick and accurate responses to inquiries from faculty, students, and other stakeholders, thereby reducing staff workload and improving customer satisfaction.

[0092] Example 1

[0093] 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."

[0094] Educational institutions require fast and accurate responses to inquiries from teachers, students, and other stakeholders, but they face a shortage of human resources to handle these needs. Furthermore, the wide variety of inquiries means that individual responses are limited, making it particularly difficult to respond at night or on holidays. To solve this problem, automated systems are needed, but typical automated systems have low accuracy, making it difficult to improve user satisfaction.

[0095] 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.

[0096] In this invention, the server includes a means for a user to send a query to an API endpoint using a terminal, a means for the server to analyze the received query, extract the user ID and the query content, and record them in a log file, a means for the server to generate a response to the query based on past inquiry data using a generative AI model, and a means for returning the generated response to the user's terminal in JSON format, thereby reducing the burden on staff at the educational institution and enabling quick and accurate responses.

[0097] "User" refers to a teacher, student or other party of an educational institution who accesses the system and submits a query.

[0098] "Device" refers to electronic devices used by users, such as smartphones, PCs, and tablets.

[0099] "API endpoint" refers to a server-side interface through which users can submit queries.

[0100] A "query" refers to a question or inquiry that a user poses to a system.

[0101] "Server" refers to a central device for receiving queries from users, analyzing them, and generating responses.

[0102] "User ID" refers to a unique identifier used to identify a user on the system.

[0103] "JSON format" refers to a lightweight text-based data exchange format for structuring and sending data.

[0104] A "log file" is a file that records the system's operation history and events.

[0105] A "generative AI model" refers to an artificial intelligence model that self-learns based on past inquiry data and generates appropriate responses.

[0106] "Past inquiry data" refers to the history of queries and responses the system has received in the past.

[0107] "HTTP response" refers to a communication method used by a server to return a response generated in response to a user's query.

[0108] The present invention is a system for automating inquiries from teachers, students, and other stakeholders at educational institutions. The system accepts queries from users, analyzes the queries, generates appropriate responses using a generative AI model, and returns the responses in real time. A specific implementation of this system is described below.

[0109] First, a user uses a device such as a smartphone or PC to send a query to an API endpoint. This query is sent as an HTTP POST request, and the data is included in JSON format. As an example, consider the case where a user sends the following query:

[0110] json

[0111] {

[0112] "user_id": "student123",

[0113] "query": "When is the next math test?"

[0114] }

[0115] The server then receives this HTTP POST request, parses the JSON data to extract the user ID and the query, and records this information in a log file along with the current timestamp, thus preserving the complete query history for future analysis and troubleshooting.

[0116] The server uses a generative AI model to generate an appropriate response to the received query. This AI model self-learns based on past query data, so it can always provide the best response. For example, for the following query:

[0117] "Example of a prompt to input to a generative AI model: When is the next math test?"

[0118] The generative AI model generates a response like this:

[0119] The next math test is next Friday.

[0120] Finally, the server converts the generated response into JSON format and sends it back to the user's device as an HTTP response, where the user can view it and obtain the required information.

[0121] This system operates in a server environment that operates 24 hours a day, so users can make inquiries at any time and receive prompt and accurate answers. Another major advantage is that it significantly reduces the workload of educational institution staff.

[0122] The above is a specific embodiment of the present invention.

[0123] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0124] Step 1:

[0125] A user uses a terminal to send a query to an API endpoint. As input, they provide a JSON data containing the user ID and the question. For example, they send a query like this:

[0126] json

[0127] {

[0128] "user_id": "student123",

[0129] "query": "When is the next math test?"

[0130] }

[0131] This data is sent from the device to the server over the Internet as an HTTP POST request. The output is the request data that arrives at the server.

[0132] Step 2:

[0133] The server parses the JSON data of the received HTTP POST request. The input contains the received JSON data. Specifically, it extracts the user_id and query from the request body.

[0134] The output is the parsed data (user ID and query content). The specific data processing at this stage involves parsing the JSON data and extracting fields.

[0135] Step 3:

[0136] The server records the extracted user ID and query content along with the current timestamp in a log file. The input contains the parsed data (user ID, query content, timestamp). The specific operation is to format this information and write it to a log file.

[0137] The output is a successful logging event. This process also persists data and maintains history.

[0138] Step 4:

[0139] The server calls the function generate_response() to generate an appropriate response to the input query. The input includes the parsed query content. The generative AI model self-trains based on past inquiry data to generate highly accurate answers.

[0140] For example, the following text is output in response to the query "When is the next math test?":

[0141] The next math test is next Friday.

[0142] Specifically, the operation involves inputting a query into an AI model and generating a response.

[0143] Step 5:

[0144] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response. The input includes the response text from the generative AI model. The specific operation is to encode the response text into JSON format and send it to the user's device using a communication protocol.

[0145] The output is the response data that can be viewed on the user's terminal. This process involves formatting and sending the data.

[0146] Step 6:

[0147] The user sees the response from the server on the device. The input includes the response data sent from the server in JSON format. The specific operation is that the device parses the response and displays it to the user in a human-readable format.

[0148] The output is a response that the user can see, allowing the user to get a quick and accurate answer.

[0149] (Application example 1)

[0150] 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."

[0151] Responding to inquiries at educational institutions and brick-and-mortar stores requires a great deal of effort and manpower. Responding to inquiries outside of business hours is particularly difficult at brick-and-mortar stores, posing a risk of damaging customer satisfaction. Furthermore, the wide variety of inquiries makes it difficult to provide consistent responses, increasing the burden on employees. In these circumstances, there is a need for an efficient and accurate automated response system.

[0152] 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.

[0153] In this invention, the server includes means for generating responses to user queries using an AI model, means for recording received user queries and user IDs, means for generating responses to the received queries and returning the responses in JSON format, and means for automating responses to customer inquiries in physical stores. This enables effective and efficient response to inquiries in educational institutions and physical stores. Furthermore, a server environment that can operate 24 hours a day allows responses even outside of business hours, improving customer satisfaction.

[0154] An "AI model" is a mathematical model that uses artificial intelligence to accomplish a specific task.

[0155] A "query" is an inquiry made by a user to a system, and refers to a specific question or request.

[0156] A "response" is an answer that an AI model generates based on a query, and refers to the information it provides in response to a user's question.

[0157] A "user ID" is a unique identifier for identifying a specific user and is used to manage user information.

[0158] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and representing data.

[0159] A "brick and mortar store" is a commercial establishment that offers goods or services in a physical location.

[0160] "Automation" refers to the process of manually performing a task automatically using machines or software.

[0161] A "server environment" is a collection of hardware and software that provides the computing resources and network required for a system to operate.

[0162] A "log" is data that a system uses to record various events and transactions, and is stored for later reference.

[0163] "Self-learning" refers to the learning process by which an AI model improves its performance based on past data.

[0164] This invention shows how to realize an automatic response system for customer inquiries in a physical store. Specifically, the system is constructed as follows using a server, a user terminal, and an AI model.

[0165] System configuration

[0166] Hardware and Software

[0167] Server: A server environment that can operate 24 hours a day (e.g., AWS (registered trademark) EC2 server) that runs the following software:

[0168] Server software: Python and Flask are used to provide API endpoints.

[0169] AI Model: A generative AI model built using TENSORFLOW® or PyTorch.

[0170] Log management: Use a log management library such as loguru.

[0171] User Device: A smartphone (iOS or ANDROID®) used by a customer, running the following software:

[0172] Mobile App: An application developed using React Native.

[0173] System Operation Overview

[0174] Query reception

[0175] A user enters a query using a mobile app on their smartphone and sends it as an HTTP POST request in JSON format to the server's API endpoint. This request includes the user ID and the query content.

[0176] Query Parsing and Logging

[0177] The server analyzes the user ID and query content from the received request and records this information in a log file along with the current timestamp, thus saving the query history.

[0178] Generating a response

[0179] The server calls the AI ​​model to generate an appropriate response to the query. The AI ​​model is self-trained based on past query data and is trained to provide a response appropriate to the query. The generated response is converted into JSON format.

[0180] Returning a response

[0181] The server returns the generated response to the user's device as an HTTP response, and the user can check the answer in real time on the device's application.

[0182] Specific examples

[0183] As an example, the processing flow when a customer sends a query such as "What are the opening hours of this store?" is shown below.

[0184] User submits the query:

[0185] json

[0186] {

[0187] "user_id": "customer456",

[0188] "query": "What are the opening hours for this store?"

[0189] }

[0190] Example of what the server logs:

[0191] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[0192] Response generation by AI model:

[0193] This store is open from 9am to 8pm.

[0194] Response that the server returns to the user:

[0195] json

[0196] {

[0197] "response": "This store is open from 9:00 AM to 8:00 PM."

[0198] }

[0199] Prompt Sentence Examples

[0200] Examples of specific prompts include:

[0201] Generate a response for the query: "What are this store's business hours?" based on the historical data.

[0202] This will enable stores to respond to customer inquiries quickly and accurately. In addition, a server environment that can operate 24 hours a day means customers can be served even outside of business hours, improving customer satisfaction.

[0203] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0204] Step 1:

[0205] A user inputs a query using a terminal.

[0206] Input: User ID and query content

[0207] Specific operation: The user opens the app on their smartphone, enters a question such as "What are the opening hours of this store?" and presses the send button.

[0208] Output: HTTP POST request in JSON format

[0209] json

[0210] {

[0211] "user_id": "customer456",

[0212] "query": "What are the opening hours for this store?"

[0213] }

[0214] Step 2:

[0215] The server receives the query and performs the analysis.

[0216] Input: HTTP POST request in JSON format

[0217] Specific operation: The server receives a request at an API endpoint built with Flask. It extracts the user ID and query content from the request.

[0218] Output: User ID and query content

[0219] User ID: customer456

[0220] Query: What are the opening hours for this store?

[0221] Step 3:

[0222] The server records parsed query information in a log file.

[0223] Input: User ID, query content, and timestamp

[0224] Specific operation: The server uses the Loguru library to record the received information in a log file.

[0225] Output: Recorded in a log file

[0226] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[0227] Step 4:

[0228] The server invokes the AI ​​model to generate a response to the query.

[0229] Input: Query content

[0230] How it works: The server loads a TensorFlow or PyTorch model and inputs the query content into the AI ​​model, which generates a response based on the trained data.

[0231] Output: The generated response

[0232] This store is open from 9am to 8pm.

[0233] Step 5:

[0234] The server converts the generated response into JSON format and sends it back to the user device.

[0235] Input: Generated response

[0236] What happens: The server uses Python's JSON library to serialize the response into JSON format, which it then sends back as an HTTP response.

[0237] Output: HTTP response in JSON format

[0238] json

[0239] {

[0240] "response": "This store is open from 9:00 AM to 8:00 PM."

[0241] }

[0242] Step 6:

[0243] The user terminal receives and displays the response from the server.

[0244] Input: HTTP response in JSON format

[0245] Specific behavior: The user device receives the HTTP response, deserializes it appropriately within the app, and displays it. The user confirms the answer displayed: "This store's business hours are from 9:00 AM to 8:00 PM."

[0246] Output: The response displayed on the screen

[0247] This store is open from 9am to 8pm.

[0248] 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.

[0249] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[0250] This system has the following main functions:

[0251] 1. Receiving queries from users

[0252] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0253] 2. Query Parsing and Logging

[0254] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0255] 3. Emotional Recognition

[0256] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, if the query is "When is the next math test?", emotions such as tension and anxiety will be recognized.

[0257] 4. Generating the Response

[0258] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[0259] 5. Adjusting the response

[0260] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[0261] 6. Returning a Response

[0262] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0263] Natural language explanation of the process

[0264] Query reception

[0265] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[0266] json

[0267] {

[0268] "user_id": "student123",

[0269] "query": "When is the next math test?"

[0270] }

[0271] Query Parsing and Logging

[0272] The server receives this request and parses the user_id and query values, which are logged as follows:

[0273] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0274] Emotion recognition

[0275] The server uses an emotion engine to analyze the user's emotions from the query, for example, recognizing emotions such as "anxiety" or "tension" from the query.

[0276] Generating a response

[0277] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model, which uses the past query history and the output of the emotion engine to generate the response.

[0278] Response adjustment

[0279] The server tailors the generated response based on the user's emotions, for example, choosing reassuring words if the user is feeling anxious.

[0280] Returning a response

[0281] The server converts the response to JSON format and returns it to the user's device as an HTTP response, which the user sees on their device, looking something like this:

[0282] json

[0283] {

[0284] "response": "The next math test is next Friday. Don't worry."

[0285] }

[0286] The system allows faculty, students and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[0287] The processing flow will be explained below.

[0288] Step 1:

[0289] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[0290] json

[0291] {

[0292] "user_id": "student123",

[0293] "query": "When is the next math test?"

[0294] }

[0295] Step 2:

[0296] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[0297] Step 3:

[0298] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[0299] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0300] Step 4:

[0301] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing algorithms to identify the emotion contained in the query text. For example, it can recognize emotions such as "anxiety" or "tension" from the query "When is the next math test?"

[0302] Step 5:

[0303] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model takes the query history and the output of the emotion engine as input and generates an optimal response. Here, the response is generated taking emotions into account.

[0304] Step 6:

[0305] The server adjusts the generated response based on the user's emotions. For example, if the user is feeling anxious, it adds reassuring words to the response text. Specifically, it generates a response of the form, "The next math test is next Friday. Don't worry."

[0306] Step 7:

[0307] The server then converts this response to JSON by using the jsonify function to package the resulting text as a JSON object, resulting in a JSON response like this:

[0308] json

[0309] {

[0310] "response": "The next math test is next Friday. Don't worry."

[0311] }

[0312] Step 8:

[0313] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[0314] AI chatbot replies: The next math test is next Friday. Don't worry.

[0315] The above processing steps enable the provision of fast and accurate responses to inquiries from faculty, students, and other stakeholders of educational institutions. Responses that take emotion into consideration can provide users with more satisfactory service.

[0316] Example 2

[0317] 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."

[0318] Educational institutions need to respond quickly and appropriately to inquiries from faculty, students, and other stakeholders, but doing so manually is labor-intensive and time-consuming, and the quality of responses varies. Furthermore, there is a need for systems that can respond appropriately to queries that contain emotions, but conventional systems have difficulty recognizing user emotions and generating appropriate responses based on those emotions.

[0319] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0320] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording the received user query and user ID, means for analyzing the received query to recognize the user's emotion, means for adjusting the generated response based on the recognized emotion, and means for returning the generated response in JSON format, thereby enabling the generation and return of a prompt and appropriate response that takes the user's emotion into consideration.

[0321] An "AI model" is a program that uses artificial intelligence algorithms to recognize patterns and generate responses.

[0322] A "user" is an individual or entity that uses the system to make an inquiry.

[0323] A "query" is a question or request that a user sends to the system.

[0324] "User ID" means a unique identifier used by the system to identify a user.

[0325] An "emotion engine" is a software component that uses natural language processing to determine a user's emotion from text.

[0326] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format for representing data in the form of JavaScript objects.

[0327] An "HTTP POST request" is a type of HTTP method for sending data to a web server.

[0328] A "log file" is a file that records system operations and user activities.

[0329] "Response adjustment" is the process of appropriately modifying the generated response based on the user's emotions.

[0330] An "HTTP response" is a response message that a server sends to a client.

[0331] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[0332] The system uses the following main hardware and software:

[0333] Hardware: Smartphones, PCs, servers

[0334] Software: AI model, emotion engine, logging system

[0335] The roles of each piece of hardware and software are as follows:

[0336] 1. Receiving queries

[0337] A user sends a query to the API endpoint using a device such as a smartphone or PC. This query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0338] 2. Query Parsing and Logging

[0339] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0340] 3. Emotional Recognition

[0341] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, it can recognize emotions such as "anxiety" and "tension" from the query "When is the next math test?"

[0342] 4. Generating the Response

[0343] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[0344] 5. Adjusting the response

[0345] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[0346] 6. Returning a Response

[0347] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0348] Examples of prompt statements

[0349] Here are some examples of prompts to input to a generative AI model:

[0350] Prompt statement example 1:

[0351] "When is the next math test?"

[0352] An example response to this:

[0353] "The next math test is next Friday. Don't worry."

[0354] Prompt statement example 2:

[0355] "Please tell me how to create a report card."

[0356] An example response to this:

[0357] "I will teach you how to create a report card. First, please log in to your school's portal site. Then, select the 'Report Card' tab and enter the required information. Please feel free to let me know if you have any additional questions."

[0358] This system allows faculty, students, and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[0359] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0360] Step 1: Query Reception

[0361] A user sends a query to the API endpoint using a smartphone or PC. The data sent is an HTTP POST request in JSON format. For example, the following input data is sent:

[0362] json

[0363] {

[0364] "user_id": "student123",

[0365] "query": "When is the next math test?"

[0366] }

[0367] The query is accepted when the device sends this JSON data to the API endpoint. The input includes the user ID and the query content. The output is received by the server.

[0368] Step 2: Parsing and logging queries

[0369] The server parses the incoming HTTP POST request and extracts the user_id and query values. For example, given the following input data:

[0370] json

[0371] {

[0372] "user_id": "student123",

[0373] "query": "When is the next math test?"

[0374] }

[0375] The server records the extracted user ID "student123" and the query content "When is the next math test?" along with a timestamp in a log file. Data processing here involves parsing the received data and generating log data.

[0376] As output, this will produce a log entry like this:

[0377] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0378] Step 3: Recognize emotions

[0379] The server inputs the query content into the emotion engine to recognize the user's emotion. The input in this process is the query "When is the next math test?"

[0380] The server analyzes the query with its emotion engine and recognizes the emotion, for example, "anxiety." As output, it gets the following data:

[0381] json

[0382] {

[0383] "emotion": "anxiety"

[0384] }

[0385] Specific operations include an emotion engine evaluating the query content using natural language processing algorithms.

[0386] Step 4: Generate a response

[0387] The server uses an AI model to generate a response based on the query and the perceived emotion. In this process, the query "When is the next math test?" and the emotion "anxiety" are input.

[0388] The server uses the function generate_response() to generate a response like this:

[0389] json

[0390] {

[0391] "response": "The next math test is next Friday. Don't worry."

[0392] }

[0393] The data calculation here is the generation of response text by an AI model.

[0394] Step 5: Adjust the response

[0395] The server adjusts the generated response based on the recognized emotion. Since the emotion is "anxious," the response includes a reassuring expression. The inputs are the generated response and the recognized emotion.

[0396] For example, the original response, "The next math test is next Friday," is adjusted to "The next math test is next Friday. Don't worry." The output is the adjusted response.

[0397] Step 6: Returning the response

[0398] The server then converts the final adjusted response into JSON format and sends it to the user's device as an HTTP response, with the adjusted response text as input.

[0399] json

[0400] {

[0401] "response": "The next math test is next Friday. Don't worry."

[0402] }

[0403] The server sends this JSON data to the user's device, where they can view it in real time. The output is an HTTP response containing the response.

[0404] (Application example 2)

[0405] 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."

[0406] Conventional customer support systems have the problem of being unable to properly grasp user emotions, resulting in inappropriate responses and a decline in user satisfaction. Furthermore, since it is difficult to provide 24-hour support, delays in response can lead to a loss of user trust. In the security field, in particular, it is necessary to quickly alleviate user anxiety and tension, so these issues must be resolved.

[0407] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0408] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording received user queries and user IDs, means for generating a response to the received query and returning the response in JSON format, means for analyzing the user's emotion from the query using an emotion engine, and means for adjusting the generated response based on the user's emotion, thereby making it possible to provide appropriate responses according to the user's emotion 24 hours a day.

[0409] "AI model" refers to artificial intelligence that generates appropriate responses to user queries.

[0410] A "query" refers to a question or request that a user makes to a system.

[0411] "User ID" means a unique identifier used to identify a User.

[0412] "JSON format" stands for JavaScript Object Notation and refers to a way of representing data in a lightweight text format.

[0413] An "emotion engine" refers to software or a system for analyzing emotions from user queries.

[0414] "Self-learning" refers to the process of learning from past inquiry history to improve the accuracy and adaptability of responses.

[0415] "Server environment" refers to the hardware and software infrastructure on which the system operates.

[0416] The system of the present invention generates appropriate responses to user queries, adjusts the responses based on the user's emotions, and provides them on a 24-hour basis. A specific embodiment of the system is described below.

[0417] The system mainly consists of a server, a user terminal and related software.

[0418] System Configuration

[0419] 1. Server

[0420] The server is a computer with a powerful CPU and sufficient memory that receives, processes, and transmits data. Software used includes Python 3, Flask (or Django), etc. The server has the following features:

[0421] AI Model: Used to generate appropriate responses to user queries.

[0422] Emotion Engine: Uses natural language processing techniques to analyze user emotions from queries.

[0423] 2. User Device

[0424] A user device is a smartphone, computer, or other device used to send questions or requests. Users use their device to send queries to API endpoints.

[0425] Processing flow

[0426] 1. Receiving queries from users

[0427] A user uses a device to send a query to an API endpoint. The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0428] Examples:

[0429] Examples of data users submit:

[0430] User ID: student123

[0431] Query: When is the next math test?

[0432] 2. Query Parsing and Logging

[0433] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0434] 3. Emotional Recognition

[0435] The server analyzes the user's emotions from the query using an emotion engine, which uses natural language processing techniques to identify the emotions contained in the query content.

[0436] Examples:

[0437] Emotions recognized by the emotion engine: anxiety, tension, anger, etc.

[0438] Query:

[0439] User ID: student123

[0440] Query: When is the next math test?

[0441] Recognized emotion: Anxiety

[0442] 4. Generating the Response

[0443] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[0444] Examples:

[0445] Response to the query:

[0446] The next math test is next Friday.

[0447] 5. Adjusting the response

[0448] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[0449] Examples:

[0450] Adjusted response:

[0451] "The next math test is next Friday. Don't worry."

[0452] 6. Returning a Response

[0453] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0454] Examples:

[0455] The final response returned to the user:

[0456] "The next math test is next Friday. Don't worry."

[0457] Prompt Sentence Examples

[0458] User ID: user456

[0459] Query: Tell us about any recent security incidents.

[0460] Emotion: Anxiety

[0461] As described above, the system of the present invention is capable of providing appropriate responses based on the user's emotions 24 hours a day, thereby improving user satisfaction and reliability.

[0462] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0463] Step 1:

[0464] A user uses a device to send a query to an API endpoint. As input, the device sends JSON data containing the user ID and the query content. The device then sends this data as an HTTP POST request.

[0465] Specific behavior: The user enters a question into the application and presses the "Submit" button to send the data.

[0466] Step 2:

[0467] The server analyzes the request received. It receives a JSON-formatted query as input, extracts the user ID and the query content from it, and records this information and the current timestamp in a log file.

[0468] Specific behavior: The server parses the received data and logs it as follows. For example: "Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?"

[0469] Step 3:

[0470] The server uses an emotion engine to analyze the user's emotion from the query. Using the query content as input, the emotion engine identifies the emotion using natural language processing techniques. The identified emotion is obtained as output.

[0471] Specific behavior: The emotion engine identifies the emotion "anxiety" for the query "When is the next math test?"

[0472] Step 4:

[0473] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The input is the query and the analyzed emotion, and the AI ​​model generates a response based on the past query history and the emotion. The output is the generated response.

[0474] How it works: The AI ​​model generates the response "The next math test is next Friday" based on past data and the emotion "anxiety."

[0475] Step 5:

[0476] The server adjusts the generated response based on the user's emotion. The inputs are the generated response and the user's emotion, and the server takes into account the output of the emotion engine to adjust the response. The output is the adjusted response.

[0477] Behavior: Adjust the generated response "The next math test is next Friday" by adding the phrase "Don't worry, it's okay."

[0478] Step 6:

[0479] The server converts the adjusted response to JSON format and sends it to the user's device as an HTTP response. The input is the adjusted response, which the server converts to JSON format and sends to the user. The output is the response that is displayed on the user's device.

[0480] Specific behavior: In response to a question submitted by the user, the device displays a response saying, "The next math test is next Friday. Don't worry."

[0481] 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.

[0482] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0483] 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.

[0484] [Second embodiment]

[0485] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0486] 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.

[0487] 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).

[0488] 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.

[0489] 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.

[0490] 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).

[0491] 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.

[0492] 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.

[0493] 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.

[0494] 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.

[0495] In the smart glasses 214, 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.

[0496] 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."

[0497] The present invention is a system that uses AI models to automate customer support for inquiries from faculty, students, and other stakeholders at educational institutions. The system includes the following main functions:

[0498] 1. Receiving queries from users

[0499] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0500] 2. Query Parsing and Logging

[0501] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0502] 3. Generating a Response

[0503] The server generates appropriate responses to user queries using an AI model that is self-learning based on past query history and is constantly tuned to provide the best possible response.

[0504] 4. Returning a Response

[0505] The generated response is converted to JSON format and sent to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0506] Natural language explanation of the process

[0507] Query reception

[0508] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[0509] json

[0510] {

[0511] "user_id": "student123",

[0512] "query": "When is the next math test?"

[0513] }

[0514] Query Parsing and Logging

[0515] The server receives this request and parses the user_id and query values, which are logged as follows:

[0516] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0517] Generating a response

[0518] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model is actually self-trained based on past query data, which improves the accuracy of the answer given to the user. An example response might be the following text:

[0519] The next math test is next Friday.

[0520] Returning a response

[0521] The server converts the generated response into JSON format and returns it to the user's device as an HTTP response, which the user sees on their device and sees something like this:

[0522] json

[0523] {

[0524] "response": "The next math test is next Friday."

[0525] }

[0526] The system operates in a 24-hour server environment, allowing users to make inquiries at any time, reducing the burden on educational institution staff and providing quick and accurate responses.

[0527] The processing flow will be explained below.

[0528] Step 1:

[0529] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[0530] json

[0531] {

[0532] "user_id": "student123",

[0533] "query": "When is the next math test?"

[0534] }

[0535] Step 2:

[0536] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[0537] Step 3:

[0538] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[0539] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0540] Step 4:

[0541] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. This function uses a self-trained AI model based on past query history to return the best possible response. In this simple example, the query is returned as the response.

[0542] Step 5:

[0543] The server converts the generated response into JSON format by using the jsonify function to package the generated text as a JSON object, for example generating a JSON response like this:

[0544] json

[0545] {

[0546] "response": "The next math test is next Friday."

[0547] }

[0548] Step 6:

[0549] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[0550] AI chatbot response: The next math test is next Friday.

[0551] The above processing steps enable the institution to provide quick and accurate responses to inquiries from faculty, students, and other stakeholders, thereby reducing staff workload and improving customer satisfaction.

[0552] Example 1

[0553] 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."

[0554] Educational institutions require fast and accurate responses to inquiries from teachers, students, and other stakeholders, but they face a shortage of human resources to handle these needs. Furthermore, the wide variety of inquiries means that individual responses are limited, making it particularly difficult to respond at night or on holidays. To solve this problem, automated systems are needed, but typical automated systems have low accuracy, making it difficult to improve user satisfaction.

[0555] 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.

[0556] In this invention, the server includes a means for a user to send a query to an API endpoint using a terminal, a means for the server to analyze the received query, extract the user ID and the query content, and record them in a log file, a means for the server to generate a response to the query based on past inquiry data using a generative AI model, and a means for returning the generated response to the user's terminal in JSON format, thereby reducing the burden on staff at the educational institution and enabling quick and accurate responses.

[0557] "User" refers to a teacher, student or other party of an educational institution who accesses the system and submits a query.

[0558] "Device" refers to electronic devices used by users, such as smartphones, PCs, and tablets.

[0559] "API endpoint" refers to a server-side interface through which users can submit queries.

[0560] A "query" refers to a question or inquiry that a user poses to a system.

[0561] "Server" refers to a central device for receiving queries from users, analyzing them, and generating responses.

[0562] "User ID" refers to a unique identifier used to identify a user on the system.

[0563] "JSON format" refers to a lightweight text-based data exchange format for structuring and sending data.

[0564] A "log file" is a file that records the system's operation history and events.

[0565] A "generative AI model" refers to an artificial intelligence model that self-learns based on past inquiry data and generates appropriate responses.

[0566] "Past inquiry data" refers to the history of queries and responses the system has received in the past.

[0567] "HTTP response" refers to a communication method used by a server to return a response generated in response to a user's query.

[0568] The present invention is a system for automating inquiries from teachers, students, and other stakeholders at educational institutions. The system accepts queries from users, analyzes the queries, generates appropriate responses using a generative AI model, and returns the responses in real time. A specific implementation of this system is described below.

[0569] First, a user uses a device such as a smartphone or PC to send a query to an API endpoint. This query is sent as an HTTP POST request, and the data is included in JSON format. As an example, consider the case where a user sends the following query:

[0570] json

[0571] {

[0572] "user_id": "student123",

[0573] "query": "When is the next math test?"

[0574] }

[0575] The server then receives this HTTP POST request, parses the JSON data to extract the user ID and the query, and records this information in a log file along with the current timestamp, thus preserving the complete query history for future analysis and troubleshooting.

[0576] The server uses a generative AI model to generate an appropriate response to the received query. This AI model self-learns based on past query data, so it can always provide the best response. For example, for the following query:

[0577] "Example of a prompt to input to a generative AI model: When is the next math test?"

[0578] The generative AI model generates a response like this:

[0579] The next math test is next Friday.

[0580] Finally, the server converts the generated response into JSON format and sends it back to the user's device as an HTTP response, where the user can view it and obtain the required information.

[0581] This system operates in a server environment that operates 24 hours a day, so users can make inquiries at any time and receive prompt and accurate answers. Another major advantage is that it significantly reduces the workload of educational institution staff.

[0582] The above is a specific embodiment of the present invention.

[0583] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0584] Step 1:

[0585] A user uses a terminal to send a query to an API endpoint. As input, they provide a JSON data containing the user ID and the question. For example, they send a query like this:

[0586] json

[0587] {

[0588] "user_id": "student123",

[0589] "query": "When is the next math test?"

[0590] }

[0591] This data is sent from the device to the server over the Internet as an HTTP POST request. The output is the request data that arrives at the server.

[0592] Step 2:

[0593] The server parses the JSON data of the received HTTP POST request. The input contains the received JSON data. Specifically, it extracts the user_id and query from the request body.

[0594] The output is the parsed data (user ID and query content). The specific data processing at this stage involves parsing the JSON data and extracting fields.

[0595] Step 3:

[0596] The server records the extracted user ID and query content along with the current timestamp in a log file. The input contains the parsed data (user ID, query content, timestamp). The specific operation is to format this information and write it to a log file.

[0597] The output is a successful logging event. This process also persists data and maintains history.

[0598] Step 4:

[0599] The server calls the function generate_response() to generate an appropriate response to the input query. The input includes the parsed query content. The generative AI model self-trains based on past inquiry data to generate highly accurate answers.

[0600] For example, the following text is output in response to the query "When is the next math test?":

[0601] The next math test is next Friday.

[0602] Specifically, the operation involves inputting a query into an AI model and generating a response.

[0603] Step 5:

[0604] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response. The input includes the response text from the generative AI model. The specific operation is to encode the response text into JSON format and send it to the user's device using a communication protocol.

[0605] The output is the response data that can be viewed on the user's terminal. This process involves formatting and sending the data.

[0606] Step 6:

[0607] The user sees the response from the server on the device. The input includes the response data sent from the server in JSON format. The specific operation is that the device parses the response and displays it to the user in a human-readable format.

[0608] The output is a response that the user can see, allowing the user to get a quick and accurate answer.

[0609] (Application example 1)

[0610] 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."

[0611] Responding to inquiries at educational institutions and brick-and-mortar stores requires a great deal of effort and manpower. Responding to inquiries outside of business hours is particularly difficult at brick-and-mortar stores, posing a risk of damaging customer satisfaction. Furthermore, the wide variety of inquiries makes it difficult to provide consistent responses, increasing the burden on employees. In these circumstances, there is a need for an efficient and accurate automated response system.

[0612] 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.

[0613] In this invention, the server includes means for generating responses to user queries using an AI model, means for recording received user queries and user IDs, means for generating responses to the received queries and returning the responses in JSON format, and means for automating responses to customer inquiries in physical stores. This enables effective and efficient response to inquiries in educational institutions and physical stores. Furthermore, a server environment that can operate 24 hours a day allows responses even outside of business hours, improving customer satisfaction.

[0614] An "AI model" is a mathematical model that uses artificial intelligence to accomplish a specific task.

[0615] A "query" is an inquiry made by a user to a system, and refers to a specific question or request.

[0616] A "response" is an answer that an AI model generates based on a query, and refers to the information it provides in response to a user's question.

[0617] A "user ID" is a unique identifier for identifying a specific user and is used to manage user information.

[0618] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and representing data.

[0619] A "brick and mortar store" is a commercial establishment that offers goods or services in a physical location.

[0620] "Automation" refers to the process of manually performing a task automatically using machines or software.

[0621] A "server environment" is a collection of hardware and software that provides the computing resources and network required for a system to operate.

[0622] A "log" is data that a system uses to record various events and transactions, and is stored for later reference.

[0623] "Self-learning" refers to the learning process by which an AI model improves its performance based on past data.

[0624] This invention shows how to realize an automatic response system for customer inquiries in a physical store. Specifically, the system is constructed as follows using a server, a user terminal, and an AI model.

[0625] System configuration

[0626] Hardware and Software

[0627] Server: A server environment that can operate 24 hours a day (e.g., AWS EC2 server) that runs the following software:

[0628] Server software: Python and Flask are used to provide API endpoints.

[0629] AI Model: A generative AI model built using TensorFlow or PyTorch.

[0630] Log management: Use a log management library such as loguru.

[0631] User Device: A smartphone (iOS or Android) used by a customer that runs the following software:

[0632] Mobile App: An application developed using React Native.

[0633] System Operation Overview

[0634] Query reception

[0635] A user enters a query using a mobile app on their smartphone and sends it as an HTTP POST request in JSON format to the server's API endpoint. This request includes the user ID and the query content.

[0636] Query Parsing and Logging

[0637] The server analyzes the user ID and query content from the received request and records this information in a log file along with the current timestamp, thus saving the query history.

[0638] Generating a response

[0639] The server calls the AI ​​model to generate an appropriate response to the query. The AI ​​model is self-trained based on past query data and is trained to provide a response appropriate to the query. The generated response is converted into JSON format.

[0640] Returning a response

[0641] The server returns the generated response to the user's device as an HTTP response, and the user can check the answer in real time on the device's application.

[0642] Specific examples

[0643] As an example, the processing flow when a customer sends a query such as "What are the opening hours of this store?" is shown below.

[0644] User submits the query:

[0645] json

[0646] {

[0647] "user_id": "customer456",

[0648] "query": "What are the opening hours for this store?"

[0649] }

[0650] Example of what the server logs:

[0651] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[0652] Response generation by AI model:

[0653] This store is open from 9am to 8pm.

[0654] Response that the server returns to the user:

[0655] json

[0656] {

[0657] "response": "This store is open from 9:00 AM to 8:00 PM."

[0658] }

[0659] Prompt Sentence Examples

[0660] Examples of specific prompts include:

[0661] Generate a response for the query: "What are this store's business hours?" based on the historical data.

[0662] This will enable stores to respond to customer inquiries quickly and accurately. In addition, a server environment that can operate 24 hours a day means customers can be served even outside of business hours, improving customer satisfaction.

[0663] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0664] Step 1:

[0665] A user inputs a query using a terminal.

[0666] Input: User ID and query content

[0667] Specific operation: The user opens the app on their smartphone, enters a question such as "What are the opening hours of this store?" and presses the send button.

[0668] Output: HTTP POST request in JSON format

[0669] json

[0670] {

[0671] "user_id": "customer456",

[0672] "query": "What are the opening hours for this store?"

[0673] }

[0674] Step 2:

[0675] The server receives the query and performs the analysis.

[0676] Input: HTTP POST request in JSON format

[0677] Specific operation: The server receives a request at an API endpoint built with Flask. It extracts the user ID and query content from the request.

[0678] Output: User ID and query content

[0679] User ID: customer456

[0680] Query: What are the opening hours for this store?

[0681] Step 3:

[0682] The server records parsed query information in a log file.

[0683] Input: User ID, query content, and timestamp

[0684] Specific operation: The server uses the Loguru library to record the received information in a log file.

[0685] Output: Recorded in a log file

[0686] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[0687] Step 4:

[0688] The server invokes the AI ​​model to generate a response to the query.

[0689] Input: Query content

[0690] How it works: The server loads a TensorFlow or PyTorch model and inputs the query content into the AI ​​model, which generates a response based on the trained data.

[0691] Output: The generated response

[0692] This store is open from 9am to 8pm.

[0693] Step 5:

[0694] The server converts the generated response into JSON format and sends it back to the user device.

[0695] Input: Generated response

[0696] What happens: The server uses Python's JSON library to serialize the response into JSON format, which it then sends back as an HTTP response.

[0697] Output: HTTP response in JSON format

[0698] json

[0699] {

[0700] "response": "This store is open from 9:00 AM to 8:00 PM."

[0701] }

[0702] Step 6:

[0703] The user terminal receives and displays the response from the server.

[0704] Input: HTTP response in JSON format

[0705] Specific behavior: The user device receives the HTTP response, deserializes it appropriately within the app, and displays it. The user confirms the answer displayed: "This store's business hours are from 9:00 AM to 8:00 PM."

[0706] Output: The response displayed on the screen

[0707] This store is open from 9am to 8pm.

[0708] 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.

[0709] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[0710] This system has the following main functions:

[0711] 1. Receiving queries from users

[0712] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0713] 2. Query Parsing and Logging

[0714] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0715] 3. Emotional Recognition

[0716] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, if the query is "When is the next math test?", emotions such as tension and anxiety will be recognized.

[0717] 4. Generating the Response

[0718] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[0719] 5. Adjusting the response

[0720] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[0721] 6. Returning a Response

[0722] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0723] Natural language explanation of the process

[0724] Query reception

[0725] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[0726] json

[0727] {

[0728] "user_id": "student123",

[0729] "query": "When is the next math test?"

[0730] }

[0731] Query Parsing and Logging

[0732] The server receives this request and parses the user_id and query values, which are logged as follows:

[0733] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0734] Emotion recognition

[0735] The server uses an emotion engine to analyze the user's emotions from the query, for example, recognizing emotions such as "anxiety" or "tension" from the query.

[0736] Generating a response

[0737] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model, which uses the past query history and the output of the emotion engine to generate the response.

[0738] Response adjustment

[0739] The server tailors the generated response based on the user's emotions, for example, choosing reassuring words if the user is feeling anxious.

[0740] Returning a response

[0741] The server converts the response to JSON format and returns it to the user's device as an HTTP response, which the user sees on their device, looking something like this:

[0742] json

[0743] {

[0744] "response": "The next math test is next Friday. Don't worry."

[0745] }

[0746] The system allows faculty, students and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[0747] The processing flow will be explained below.

[0748] Step 1:

[0749] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[0750] json

[0751] {

[0752] "user_id": "student123",

[0753] "query": "When is the next math test?"

[0754] }

[0755] Step 2:

[0756] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[0757] Step 3:

[0758] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[0759] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0760] Step 4:

[0761] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing algorithms to identify the emotion contained in the query text. For example, it can recognize emotions such as "anxiety" or "tension" from the query "When is the next math test?"

[0762] Step 5:

[0763] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model takes the query history and the output of the emotion engine as input and generates an optimal response. Here, the response is generated taking emotions into account.

[0764] Step 6:

[0765] The server adjusts the generated response based on the user's emotions. For example, if the user is feeling anxious, it adds reassuring words to the response text. Specifically, it generates a response of the form, "The next math test is next Friday. Don't worry."

[0766] Step 7:

[0767] The server then converts this response to JSON by using the jsonify function to package the resulting text as a JSON object, resulting in a JSON response like this:

[0768] json

[0769] {

[0770] "response": "The next math test is next Friday. Don't worry."

[0771] }

[0772] Step 8:

[0773] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[0774] AI chatbot replies: The next math test is next Friday. Don't worry.

[0775] The above processing steps enable the provision of fast and accurate responses to inquiries from faculty, students, and other stakeholders of educational institutions. Responses that take emotion into consideration can provide users with more satisfactory service.

[0776] Example 2

[0777] 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."

[0778] Educational institutions need to respond quickly and appropriately to inquiries from faculty, students, and other stakeholders, but doing so manually is labor-intensive and time-consuming, and the quality of responses varies. Furthermore, there is a need for systems that can respond appropriately to queries that contain emotions, but conventional systems have difficulty recognizing user emotions and generating appropriate responses based on those emotions.

[0779] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0780] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording the received user query and user ID, means for analyzing the received query to recognize the user's emotion, means for adjusting the generated response based on the recognized emotion, and means for returning the generated response in JSON format, thereby enabling the generation and return of a prompt and appropriate response that takes the user's emotion into consideration.

[0781] An "AI model" is a program that uses artificial intelligence algorithms to recognize patterns and generate responses.

[0782] A "user" is an individual or entity that uses the system to make an inquiry.

[0783] A "query" is a question or request that a user sends to the system.

[0784] "User ID" means a unique identifier used by the system to identify a user.

[0785] An "emotion engine" is a software component that uses natural language processing to determine a user's emotion from text.

[0786] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format for representing data in the form of JavaScript objects.

[0787] An "HTTP POST request" is a type of HTTP method for sending data to a web server.

[0788] A "log file" is a file that records system operations and user activities.

[0789] "Response adjustment" is the process of appropriately modifying the generated response based on the user's emotions.

[0790] An "HTTP response" is a response message that a server sends to a client.

[0791] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[0792] The system uses the following main hardware and software:

[0793] Hardware: Smartphones, PCs, servers

[0794] Software: AI model, emotion engine, logging system

[0795] The roles of each piece of hardware and software are as follows:

[0796] 1. Receiving queries

[0797] A user sends a query to the API endpoint using a device such as a smartphone or PC. This query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0798] 2. Query Parsing and Logging

[0799] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0800] 3. Emotional Recognition

[0801] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, it can recognize emotions such as "anxiety" and "tension" from the query "When is the next math test?"

[0802] 4. Generating the Response

[0803] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[0804] 5. Adjusting the response

[0805] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[0806] 6. Returning a Response

[0807] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0808] Examples of prompt statements

[0809] Here are some examples of prompts to input to a generative AI model:

[0810] Prompt statement example 1:

[0811] "When is the next math test?"

[0812] An example response to this:

[0813] "The next math test is next Friday. Don't worry."

[0814] Prompt statement example 2:

[0815] "Please tell me how to create a report card."

[0816] An example response to this:

[0817] "I will teach you how to create a report card. First, please log in to your school's portal site. Then, select the 'Report Card' tab and enter the required information. Please feel free to let me know if you have any additional questions."

[0818] This system allows faculty, students, and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[0819] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0820] Step 1: Query Reception

[0821] A user sends a query to the API endpoint using a smartphone or PC. The data sent is an HTTP POST request in JSON format. For example, the following input data is sent:

[0822] json

[0823] {

[0824] "user_id": "student123",

[0825] "query": "When is the next math test?"

[0826] }

[0827] The query is accepted when the device sends this JSON data to the API endpoint. The input includes the user ID and the query content. The output is received by the server.

[0828] Step 2: Parsing and logging queries

[0829] The server parses the incoming HTTP POST request and extracts the user_id and query values. For example, given the following input data:

[0830] json

[0831] {

[0832] "user_id": "student123",

[0833] "query": "When is the next math test?"

[0834] }

[0835] The server records the extracted user ID "student123" and the query content "When is the next math test?" along with a timestamp in a log file. Data processing here involves parsing the received data and generating log data.

[0836] As output, this will produce a log entry like this:

[0837] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0838] Step 3: Recognize emotions

[0839] The server inputs the query content into the emotion engine to recognize the user's emotion. The input in this process is the query "When is the next math test?"

[0840] The server analyzes the query with its emotion engine and recognizes the emotion, for example, "anxiety." As output, it gets the following data:

[0841] json

[0842] {

[0843] "emotion": "anxiety"

[0844] }

[0845] Specific operations include an emotion engine evaluating the query content using natural language processing algorithms.

[0846] Step 4: Generate a response

[0847] The server uses an AI model to generate a response based on the query and the perceived emotion. In this process, the query "When is the next math test?" and the emotion "anxiety" are input.

[0848] The server uses the function generate_response() to generate a response like this:

[0849] json

[0850] {

[0851] "response": "The next math test is next Friday. Don't worry."

[0852] }

[0853] The data calculation here is the generation of response text by an AI model.

[0854] Step 5: Adjust the response

[0855] The server adjusts the generated response based on the recognized emotion. Since the emotion is "anxious," the response includes a reassuring expression. The inputs are the generated response and the recognized emotion.

[0856] For example, the original response, "The next math test is next Friday," is adjusted to "The next math test is next Friday. Don't worry." The output is the adjusted response.

[0857] Step 6: Returning the response

[0858] The server then converts the final adjusted response into JSON format and sends it to the user's device as an HTTP response, with the adjusted response text as input.

[0859] json

[0860] {

[0861] "response": "The next math test is next Friday. Don't worry."

[0862] }

[0863] The server sends this JSON data to the user's device, where they can view it in real time. The output is an HTTP response containing the response.

[0864] (Application example 2)

[0865] 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."

[0866] Conventional customer support systems have the problem of being unable to properly grasp user emotions, resulting in inappropriate responses and a decline in user satisfaction. Furthermore, since it is difficult to provide 24-hour support, delays in response can lead to a loss of user trust. In the security field, in particular, it is necessary to quickly alleviate user anxiety and tension, so these issues must be resolved.

[0867] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0868] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording received user queries and user IDs, means for generating a response to the received query and returning the response in JSON format, means for analyzing the user's emotion from the query using an emotion engine, and means for adjusting the generated response based on the user's emotion, thereby making it possible to provide appropriate responses according to the user's emotion 24 hours a day.

[0869] "AI model" refers to artificial intelligence that generates appropriate responses to user queries.

[0870] A "query" refers to a question or request that a user makes to a system.

[0871] "User ID" means a unique identifier used to identify a User.

[0872] "JSON format" stands for JavaScript Object Notation and refers to a way of representing data in a lightweight text format.

[0873] An "emotion engine" refers to software or a system for analyzing emotions from user queries.

[0874] "Self-learning" refers to the process of learning from past inquiry history to improve the accuracy and adaptability of responses.

[0875] "Server environment" refers to the hardware and software infrastructure on which the system operates.

[0876] The system of the present invention generates appropriate responses to user queries, adjusts the responses based on the user's emotions, and provides them on a 24-hour basis. A specific embodiment of the system is described below.

[0877] The system mainly consists of a server, a user terminal and related software.

[0878] System Configuration

[0879] 1. Server

[0880] The server is a computer with a powerful CPU and sufficient memory that receives, processes, and transmits data. Software used includes Python 3, Flask (or Django), etc. The server has the following features:

[0881] AI Model: Used to generate appropriate responses to user queries.

[0882] Emotion Engine: Uses natural language processing techniques to analyze user emotions from queries.

[0883] 2. User Device

[0884] A user device is a smartphone, computer, or other device used to send questions or requests. Users use their device to send queries to API endpoints.

[0885] Processing flow

[0886] 1. Receiving queries from users

[0887] A user uses a device to send a query to an API endpoint. The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0888] Examples:

[0889] Examples of data users submit:

[0890] User ID: student123

[0891] Query: When is the next math test?

[0892] 2. Query Parsing and Logging

[0893] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0894] 3. Emotional Recognition

[0895] The server analyzes the user's emotions from the query using an emotion engine, which uses natural language processing techniques to identify the emotions contained in the query content.

[0896] Examples:

[0897] Emotions recognized by the emotion engine: anxiety, tension, anger, etc.

[0898] Query:

[0899] User ID: student123

[0900] Query: When is the next math test?

[0901] Recognized emotion: Anxiety

[0902] 4. Generating the Response

[0903] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[0904] Examples:

[0905] Response to the query:

[0906] The next math test is next Friday.

[0907] 5. Adjusting the response

[0908] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[0909] Examples:

[0910] Adjusted response:

[0911] "The next math test is next Friday. Don't worry."

[0912] 6. Returning a Response

[0913] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0914] Examples:

[0915] The final response returned to the user:

[0916] "The next math test is next Friday. Don't worry."

[0917] Prompt Sentence Examples

[0918] User ID: user456

[0919] Query: Tell us about any recent security incidents.

[0920] Emotion: Anxiety

[0921] As described above, the system of the present invention is capable of providing appropriate responses based on the user's emotions 24 hours a day, thereby improving user satisfaction and reliability.

[0922] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0923] Step 1:

[0924] A user uses a device to send a query to an API endpoint. As input, the device sends JSON data containing the user ID and the query content. The device then sends this data as an HTTP POST request.

[0925] Specific behavior: The user enters a question into the application and presses the "Submit" button to send the data.

[0926] Step 2:

[0927] The server analyzes the request received. It receives a JSON-formatted query as input, extracts the user ID and the query content from it, and records this information and the current timestamp in a log file.

[0928] Specific behavior: The server parses the received data and logs it as follows. For example: "Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?"

[0929] Step 3:

[0930] The server uses an emotion engine to analyze the user's emotion from the query. Using the query content as input, the emotion engine identifies the emotion using natural language processing techniques. The identified emotion is obtained as output.

[0931] Specific behavior: The emotion engine identifies the emotion "anxiety" for the query "When is the next math test?"

[0932] Step 4:

[0933] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The input is the query and the analyzed emotion, and the AI ​​model generates a response based on the past query history and the emotion. The output is the generated response.

[0934] How it works: The AI ​​model generates the response "The next math test is next Friday" based on past data and the emotion "anxiety."

[0935] Step 5:

[0936] The server adjusts the generated response based on the user's emotion. The inputs are the generated response and the user's emotion, and the server takes into account the output of the emotion engine to adjust the response. The output is the adjusted response.

[0937] Behavior: Adjust the generated response "The next math test is next Friday" by adding the phrase "Don't worry, it's okay."

[0938] Step 6:

[0939] The server converts the adjusted response to JSON format and sends it to the user's device as an HTTP response. The input is the adjusted response, which the server converts to JSON format and sends to the user. The output is the response that is displayed on the user's device.

[0940] Specific behavior: In response to a question submitted by the user, the device displays a response saying, "The next math test is next Friday. Don't worry."

[0941] 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.

[0942] 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.

[0943] 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.

[0944] [Third embodiment]

[0945] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0946] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0947] 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).

[0948] 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.

[0949] 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.

[0950] 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).

[0951] 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.

[0952] 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.

[0953] 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.

[0954] 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.

[0955] 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.

[0956] 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."

[0957] The present invention is a system that uses AI models to automate customer support for inquiries from faculty, students, and other stakeholders at educational institutions. The system includes the following main functions:

[0958] 1. Receiving queries from users

[0959] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[0960] 2. Query Parsing and Logging

[0961] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[0962] 3. Generating a Response

[0963] The server generates appropriate responses to user queries using an AI model that is self-learning based on past query history and is constantly tuned to provide the best possible response.

[0964] 4. Returning a Response

[0965] The generated response is converted to JSON format and sent to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[0966] Natural language explanation of the process

[0967] Query reception

[0968] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[0969] json

[0970] {

[0971] "user_id": "student123",

[0972] "query": "When is the next math test?"

[0973] }

[0974] Query Parsing and Logging

[0975] The server receives this request and parses the user_id and query values, which are logged as follows:

[0976] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[0977] Generating a response

[0978] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model is actually self-trained based on past query data, which improves the accuracy of the answer given to the user. An example response might be the following text:

[0979] The next math test is next Friday.

[0980] Returning a response

[0981] The server converts the generated response into JSON format and returns it to the user's device as an HTTP response, which the user sees on their device and sees something like this:

[0982] json

[0983] {

[0984] "response": "The next math test is next Friday."

[0985] }

[0986] The system operates in a 24-hour server environment, allowing users to make inquiries at any time, reducing the burden on educational institution staff and providing quick and accurate responses.

[0987] The processing flow will be explained below.

[0988] Step 1:

[0989] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[0990] json

[0991] {

[0992] "user_id": "student123",

[0993] "query": "When is the next math test?"

[0994] }

[0995] Step 2:

[0996] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[0997] Step 3:

[0998] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[0999] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1000] Step 4:

[1001] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. This function uses a self-trained AI model based on past query history to return the best possible response. In this simple example, the query is returned as the response.

[1002] Step 5:

[1003] The server converts the generated response into JSON format by using the jsonify function to package the generated text as a JSON object, for example generating a JSON response like this:

[1004] json

[1005] {

[1006] "response": "The next math test is next Friday."

[1007] }

[1008] Step 6:

[1009] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[1010] AI chatbot response: The next math test is next Friday.

[1011] The above processing steps enable the institution to provide quick and accurate responses to inquiries from faculty, students, and other stakeholders, thereby reducing staff workload and improving customer satisfaction.

[1012] Example 1

[1013] 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."

[1014] Educational institutions require fast and accurate responses to inquiries from teachers, students, and other stakeholders, but they face a shortage of human resources to handle these needs. Furthermore, the wide variety of inquiries means that individual responses are limited, making it particularly difficult to respond at night or on holidays. To solve this problem, automated systems are needed, but typical automated systems have low accuracy, making it difficult to improve user satisfaction.

[1015] 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.

[1016] In this invention, the server includes a means for a user to send a query to an API endpoint using a terminal, a means for the server to analyze the received query, extract the user ID and the query content, and record them in a log file, a means for the server to generate a response to the query based on past inquiry data using a generative AI model, and a means for returning the generated response to the user's terminal in JSON format, thereby reducing the burden on staff at the educational institution and enabling quick and accurate responses.

[1017] "User" refers to a teacher, student or other party of an educational institution who accesses the system and submits a query.

[1018] "Device" refers to electronic devices used by users, such as smartphones, PCs, and tablets.

[1019] "API endpoint" refers to a server-side interface through which users can submit queries.

[1020] A "query" refers to a question or inquiry that a user poses to a system.

[1021] "Server" refers to a central device for receiving queries from users, analyzing them, and generating responses.

[1022] "User ID" refers to a unique identifier used to identify a user on the system.

[1023] "JSON format" refers to a lightweight text-based data exchange format for structuring and sending data.

[1024] A "log file" is a file that records the system's operation history and events.

[1025] A "generative AI model" refers to an artificial intelligence model that self-learns based on past inquiry data and generates appropriate responses.

[1026] "Past inquiry data" refers to the history of queries and responses the system has received in the past.

[1027] "HTTP response" refers to a communication method used by a server to return a response generated in response to a user's query.

[1028] The present invention is a system for automating inquiries from teachers, students, and other stakeholders at educational institutions. The system accepts queries from users, analyzes the queries, generates appropriate responses using a generative AI model, and returns the responses in real time. A specific implementation of this system is described below.

[1029] First, a user uses a device such as a smartphone or PC to send a query to an API endpoint. This query is sent as an HTTP POST request, and the data is included in JSON format. As an example, consider the case where a user sends the following query:

[1030] json

[1031] {

[1032] "user_id": "student123",

[1033] "query": "When is the next math test?"

[1034] }

[1035] The server then receives this HTTP POST request, parses the JSON data to extract the user ID and the query, and records this information in a log file along with the current timestamp, thus preserving the complete query history for future analysis and troubleshooting.

[1036] The server uses a generative AI model to generate an appropriate response to the received query. This AI model self-learns based on past query data, so it can always provide the best response. For example, for the following query:

[1037] "Example of a prompt to input to a generative AI model: When is the next math test?"

[1038] The generative AI model generates a response like this:

[1039] The next math test is next Friday.

[1040] Finally, the server converts the generated response into JSON format and sends it back to the user's device as an HTTP response, where the user can view it and obtain the required information.

[1041] This system operates in a server environment that operates 24 hours a day, so users can make inquiries at any time and receive prompt and accurate answers. Another major advantage is that it significantly reduces the workload of educational institution staff.

[1042] The above is a specific embodiment of the present invention.

[1043] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1044] Step 1:

[1045] A user uses a terminal to send a query to an API endpoint. As input, they provide a JSON data containing the user ID and the question. For example, they send a query like this:

[1046] json

[1047] {

[1048] "user_id": "student123",

[1049] "query": "When is the next math test?"

[1050] }

[1051] This data is sent from the device to the server over the Internet as an HTTP POST request. The output is the request data that arrives at the server.

[1052] Step 2:

[1053] The server parses the JSON data of the received HTTP POST request. The input contains the received JSON data. Specifically, it extracts the user_id and query from the request body.

[1054] The output is the parsed data (user ID and query content). The specific data processing at this stage involves parsing the JSON data and extracting fields.

[1055] Step 3:

[1056] The server records the extracted user ID and query content along with the current timestamp in a log file. The input contains the parsed data (user ID, query content, timestamp). The specific operation is to format this information and write it to a log file.

[1057] The output is a successful logging event. This process also persists data and maintains history.

[1058] Step 4:

[1059] The server calls the function generate_response() to generate an appropriate response to the input query. The input includes the parsed query content. The generative AI model self-trains based on past inquiry data to generate highly accurate answers.

[1060] For example, the following text is output in response to the query "When is the next math test?":

[1061] The next math test is next Friday.

[1062] Specifically, the operation involves inputting a query into an AI model and generating a response.

[1063] Step 5:

[1064] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response. The input includes the response text from the generative AI model. The specific operation is to encode the response text into JSON format and send it to the user's device using a communication protocol.

[1065] The output is the response data that can be viewed on the user's terminal. This process involves formatting and sending the data.

[1066] Step 6:

[1067] The user sees the response from the server on the device. The input includes the response data sent from the server in JSON format. The specific operation is that the device parses the response and displays it to the user in a human-readable format.

[1068] The output is a response that the user can see, allowing the user to get a quick and accurate answer.

[1069] (Application example 1)

[1070] 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."

[1071] Responding to inquiries at educational institutions and brick-and-mortar stores requires a great deal of effort and manpower. Responding to inquiries outside of business hours is particularly difficult at brick-and-mortar stores, posing a risk of damaging customer satisfaction. Furthermore, the wide variety of inquiries makes it difficult to provide consistent responses, increasing the burden on employees. In these circumstances, there is a need for an efficient and accurate automated response system.

[1072] 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.

[1073] In this invention, the server includes means for generating responses to user queries using an AI model, means for recording received user queries and user IDs, means for generating responses to the received queries and returning the responses in JSON format, and means for automating responses to customer inquiries in physical stores. This enables effective and efficient response to inquiries in educational institutions and physical stores. Furthermore, a server environment that can operate 24 hours a day allows responses even outside of business hours, improving customer satisfaction.

[1074] An "AI model" is a mathematical model that uses artificial intelligence to accomplish a specific task.

[1075] A "query" is an inquiry made by a user to a system, and refers to a specific question or request.

[1076] A "response" is an answer that an AI model generates based on a query, and refers to the information it provides in response to a user's question.

[1077] A "user ID" is a unique identifier for identifying a specific user and is used to manage user information.

[1078] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and representing data.

[1079] A "brick and mortar store" is a commercial establishment that offers goods or services in a physical location.

[1080] "Automation" refers to the process of manually performing a task automatically using machines or software.

[1081] A "server environment" is a collection of hardware and software that provides the computing resources and network required for a system to operate.

[1082] A "log" is data that a system uses to record various events and transactions, and is stored for later reference.

[1083] "Self-learning" refers to the learning process by which an AI model improves its performance based on past data.

[1084] This invention shows how to realize an automatic response system for customer inquiries in a physical store. Specifically, the system is constructed as follows using a server, a user terminal, and an AI model.

[1085] System configuration

[1086] Hardware and Software

[1087] Server: A server environment that can operate 24 hours a day (e.g., AWS EC2 server) that runs the following software:

[1088] Server software: Python and Flask are used to provide API endpoints.

[1089] AI Model: A generative AI model built using TensorFlow or PyTorch.

[1090] Log management: Use a log management library such as loguru.

[1091] User Device: A smartphone (iOS or Android) used by a customer that runs the following software:

[1092] Mobile App: An application developed using React Native.

[1093] System Operation Overview

[1094] Query reception

[1095] A user enters a query using a mobile app on their smartphone and sends it as an HTTP POST request in JSON format to the server's API endpoint. This request includes the user ID and the query content.

[1096] Query Parsing and Logging

[1097] The server analyzes the user ID and query content from the received request and records this information in a log file along with the current timestamp, thus saving the query history.

[1098] Generating a response

[1099] The server calls the AI ​​model to generate an appropriate response to the query. The AI ​​model is self-trained based on past query data and is trained to provide a response appropriate to the query. The generated response is converted into JSON format.

[1100] Returning a response

[1101] The server returns the generated response to the user's device as an HTTP response, and the user can check the answer in real time on the device's application.

[1102] Specific examples

[1103] As an example, the processing flow when a customer sends a query such as "What are the opening hours of this store?" is shown below.

[1104] User submits the query:

[1105] json

[1106] {

[1107] "user_id": "customer456",

[1108] "query": "What are the opening hours for this store?"

[1109] }

[1110] Example of what the server logs:

[1111] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[1112] Response generation by AI model:

[1113] This store is open from 9am to 8pm.

[1114] Response that the server returns to the user:

[1115] json

[1116] {

[1117] "response": "This store is open from 9:00 AM to 8:00 PM."

[1118] }

[1119] Prompt Sentence Examples

[1120] Examples of specific prompts include:

[1121] Generate a response for the query: "What are this store's business hours?" based on the historical data.

[1122] This will enable stores to respond to customer inquiries quickly and accurately. In addition, a server environment that can operate 24 hours a day means customers can be served even outside of business hours, improving customer satisfaction.

[1123] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1124] Step 1:

[1125] A user inputs a query using a terminal.

[1126] Input: User ID and query content

[1127] Specific operation: The user opens the app on their smartphone, enters a question such as "What are the opening hours of this store?" and presses the send button.

[1128] Output: HTTP POST request in JSON format

[1129] json

[1130] {

[1131] "user_id": "customer456",

[1132] "query": "What are the opening hours for this store?"

[1133] }

[1134] Step 2:

[1135] The server receives the query and performs the analysis.

[1136] Input: HTTP POST request in JSON format

[1137] Specific operation: The server receives a request at an API endpoint built with Flask. It extracts the user ID and query content from the request.

[1138] Output: User ID and query content

[1139] User ID: customer456

[1140] Query: What are the opening hours for this store?

[1141] Step 3:

[1142] The server records parsed query information in a log file.

[1143] Input: User ID, query content, and timestamp

[1144] Specific operation: The server uses the Loguru library to record the received information in a log file.

[1145] Output: Recorded in a log file

[1146] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[1147] Step 4:

[1148] The server invokes the AI ​​model to generate a response to the query.

[1149] Input: Query content

[1150] How it works: The server loads a TensorFlow or PyTorch model and inputs the query content into the AI ​​model, which generates a response based on the trained data.

[1151] Output: The generated response

[1152] This store is open from 9am to 8pm.

[1153] Step 5:

[1154] The server converts the generated response into JSON format and sends it back to the user device.

[1155] Input: Generated response

[1156] What happens: The server uses Python's JSON library to serialize the response into JSON format, which it then sends back as an HTTP response.

[1157] Output: HTTP response in JSON format

[1158] json

[1159] {

[1160] "response": "This store is open from 9:00 AM to 8:00 PM."

[1161] }

[1162] Step 6:

[1163] The user terminal receives and displays the response from the server.

[1164] Input: HTTP response in JSON format

[1165] Specific behavior: The user device receives the HTTP response, deserializes it appropriately within the app, and displays it. The user confirms the answer displayed: "This store's business hours are from 9:00 AM to 8:00 PM."

[1166] Output: The response displayed on the screen

[1167] This store is open from 9am to 8pm.

[1168] 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.

[1169] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[1170] This system has the following main functions:

[1171] 1. Receiving queries from users

[1172] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[1173] 2. Query Parsing and Logging

[1174] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[1175] 3. Emotional Recognition

[1176] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, if the query is "When is the next math test?", emotions such as tension and anxiety will be recognized.

[1177] 4. Generating the Response

[1178] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[1179] 5. Adjusting the response

[1180] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[1181] 6. Returning a Response

[1182] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[1183] Natural language explanation of the process

[1184] Query reception

[1185] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[1186] json

[1187] {

[1188] "user_id": "student123",

[1189] "query": "When is the next math test?"

[1190] }

[1191] Query Parsing and Logging

[1192] The server receives this request and parses the user_id and query values, which are logged as follows:

[1193] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1194] Emotion recognition

[1195] The server uses an emotion engine to analyze the user's emotions from the query, for example, recognizing emotions such as "anxiety" or "tension" from the query.

[1196] Generating a response

[1197] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model, which uses the past query history and the output of the emotion engine to generate the response.

[1198] Response adjustment

[1199] The server tailors the generated response based on the user's emotions, for example, choosing reassuring words if the user is feeling anxious.

[1200] Returning a response

[1201] The server converts the response to JSON format and returns it to the user's device as an HTTP response, which the user sees on their device, looking something like this:

[1202] json

[1203] {

[1204] "response": "The next math test is next Friday. Don't worry."

[1205] }

[1206] The system allows faculty, students and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[1207] The processing flow will be explained below.

[1208] Step 1:

[1209] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[1210] json

[1211] {

[1212] "user_id": "student123",

[1213] "query": "When is the next math test?"

[1214] }

[1215] Step 2:

[1216] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[1217] Step 3:

[1218] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[1219] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1220] Step 4:

[1221] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing algorithms to identify the emotion contained in the query text. For example, it can recognize emotions such as "anxiety" or "tension" from the query "When is the next math test?"

[1222] Step 5:

[1223] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model takes the query history and the output of the emotion engine as input and generates an optimal response. Here, the response is generated taking emotions into account.

[1224] Step 6:

[1225] The server adjusts the generated response based on the user's emotions. For example, if the user is feeling anxious, it adds reassuring words to the response text. Specifically, it generates a response of the form, "The next math test is next Friday. Don't worry."

[1226] Step 7:

[1227] The server then converts this response to JSON by using the jsonify function to package the resulting text as a JSON object, resulting in a JSON response like this:

[1228] json

[1229] {

[1230] "response": "The next math test is next Friday. Don't worry."

[1231] }

[1232] Step 8:

[1233] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[1234] AI chatbot replies: The next math test is next Friday. Don't worry.

[1235] The above processing steps enable the provision of fast and accurate responses to inquiries from faculty, students, and other stakeholders of educational institutions. Responses that take emotion into consideration can provide users with more satisfactory service.

[1236] Example 2

[1237] 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."

[1238] Educational institutions need to respond quickly and appropriately to inquiries from faculty, students, and other stakeholders, but doing so manually is labor-intensive and time-consuming, and the quality of responses varies. Furthermore, there is a need for systems that can respond appropriately to queries that contain emotions, but conventional systems have difficulty recognizing user emotions and generating appropriate responses based on those emotions.

[1239] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1240] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording the received user query and user ID, means for analyzing the received query to recognize the user's emotion, means for adjusting the generated response based on the recognized emotion, and means for returning the generated response in JSON format, thereby enabling the generation and return of a prompt and appropriate response that takes the user's emotion into consideration.

[1241] An "AI model" is a program that uses artificial intelligence algorithms to recognize patterns and generate responses.

[1242] A "user" is an individual or entity that uses the system to make an inquiry.

[1243] A "query" is a question or request that a user sends to the system.

[1244] "User ID" means a unique identifier used by the system to identify a user.

[1245] An "emotion engine" is a software component that uses natural language processing to determine a user's emotion from text.

[1246] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format for representing data in the form of JavaScript objects.

[1247] An "HTTP POST request" is a type of HTTP method for sending data to a web server.

[1248] A "log file" is a file that records system operations and user activities.

[1249] "Response adjustment" is the process of appropriately modifying the generated response based on the user's emotions.

[1250] An "HTTP response" is a response message that a server sends to a client.

[1251] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[1252] The system uses the following main hardware and software:

[1253] Hardware: Smartphones, PCs, servers

[1254] Software: AI model, emotion engine, logging system

[1255] The roles of each piece of hardware and software are as follows:

[1256] 1. Receiving queries

[1257] A user sends a query to the API endpoint using a device such as a smartphone or PC. This query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[1258] 2. Query Parsing and Logging

[1259] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[1260] 3. Emotional Recognition

[1261] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, it can recognize emotions such as "anxiety" and "tension" from the query "When is the next math test?"

[1262] 4. Generating the Response

[1263] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[1264] 5. Adjusting the response

[1265] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[1266] 6. Returning a Response

[1267] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[1268] Examples of prompt statements

[1269] Here are some examples of prompts to input to a generative AI model:

[1270] Prompt statement example 1:

[1271] "When is the next math test?"

[1272] An example response to this:

[1273] "The next math test is next Friday. Don't worry."

[1274] Prompt statement example 2:

[1275] "Please tell me how to create a report card."

[1276] An example response to this:

[1277] "I will teach you how to create a report card. First, please log in to your school's portal site. Then, select the 'Report Card' tab and enter the required information. Please feel free to let me know if you have any additional questions."

[1278] This system allows faculty, students, and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[1279] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1280] Step 1: Query Reception

[1281] A user sends a query to the API endpoint using a smartphone or PC. The data sent is an HTTP POST request in JSON format. For example, the following input data is sent:

[1282] json

[1283] {

[1284] "user_id": "student123",

[1285] "query": "When is the next math test?"

[1286] }

[1287] The query is accepted when the device sends this JSON data to the API endpoint. The input includes the user ID and the query content. The output is received by the server.

[1288] Step 2: Parsing and logging queries

[1289] The server parses the incoming HTTP POST request and extracts the user_id and query values. For example, given the following input data:

[1290] json

[1291] {

[1292] "user_id": "student123",

[1293] "query": "When is the next math test?"

[1294] }

[1295] The server records the extracted user ID "student123" and the query content "When is the next math test?" along with a timestamp in a log file. Data processing here involves parsing the received data and generating log data.

[1296] As output, this will produce a log entry like this:

[1297] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1298] Step 3: Recognize emotions

[1299] The server inputs the query content into the emotion engine to recognize the user's emotion. The input in this process is the query "When is the next math test?"

[1300] The server analyzes the query with its emotion engine and recognizes the emotion, for example, "anxiety." As output, it gets the following data:

[1301] json

[1302] {

[1303] "emotion": "anxiety"

[1304] }

[1305] Specific operations include an emotion engine evaluating the query content using natural language processing algorithms.

[1306] Step 4: Generate a response

[1307] The server uses an AI model to generate a response based on the query and the perceived emotion. In this process, the query "When is the next math test?" and the emotion "anxiety" are input.

[1308] The server uses the function generate_response() to generate a response like this:

[1309] json

[1310] {

[1311] "response": "The next math test is next Friday. Don't worry."

[1312] }

[1313] The data calculation here is the generation of response text by an AI model.

[1314] Step 5: Adjust the response

[1315] The server adjusts the generated response based on the recognized emotion. Since the emotion is "anxious," the response includes a reassuring expression. The inputs are the generated response and the recognized emotion.

[1316] For example, the original response, "The next math test is next Friday," is adjusted to "The next math test is next Friday. Don't worry." The output is the adjusted response.

[1317] Step 6: Returning the response

[1318] The server then converts the final adjusted response into JSON format and sends it to the user's device as an HTTP response, with the adjusted response text as input.

[1319] json

[1320] {

[1321] "response": "The next math test is next Friday. Don't worry."

[1322] }

[1323] The server sends this JSON data to the user's device, where they can view it in real time. The output is an HTTP response containing the response.

[1324] (Application example 2)

[1325] 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."

[1326] Conventional customer support systems have the problem of being unable to properly grasp user emotions, resulting in inappropriate responses and a decline in user satisfaction. Furthermore, since it is difficult to provide 24-hour support, delays in response can lead to a loss of user trust. In the security field, in particular, it is necessary to quickly alleviate user anxiety and tension, so these issues must be resolved.

[1327] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1328] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording received user queries and user IDs, means for generating a response to the received query and returning the response in JSON format, means for analyzing the user's emotion from the query using an emotion engine, and means for adjusting the generated response based on the user's emotion, thereby making it possible to provide appropriate responses according to the user's emotion 24 hours a day.

[1329] "AI model" refers to artificial intelligence that generates appropriate responses to user queries.

[1330] A "query" refers to a question or request that a user makes to a system.

[1331] "User ID" means a unique identifier used to identify a User.

[1332] "JSON format" stands for JavaScript Object Notation and refers to a way of representing data in a lightweight text format.

[1333] An "emotion engine" refers to software or a system for analyzing emotions from user queries.

[1334] "Self-learning" refers to the process of learning from past inquiry history to improve the accuracy and adaptability of responses.

[1335] "Server environment" refers to the hardware and software infrastructure on which the system operates.

[1336] The system of the present invention generates appropriate responses to user queries, adjusts the responses based on the user's emotions, and provides them on a 24-hour basis. A specific embodiment of the system is described below.

[1337] The system mainly consists of a server, a user terminal and related software.

[1338] System Configuration

[1339] 1. Server

[1340] The server is a computer with a powerful CPU and sufficient memory that receives, processes, and transmits data. Software used includes Python 3, Flask (or Django), etc. The server has the following features:

[1341] AI Model: Used to generate appropriate responses to user queries.

[1342] Emotion Engine: Uses natural language processing techniques to analyze user emotions from queries.

[1343] 2. User Device

[1344] A user device is a smartphone, computer, or other device used to send questions or requests. Users use their device to send queries to API endpoints.

[1345] Processing flow

[1346] 1. Receiving queries from users

[1347] A user uses a device to send a query to an API endpoint. The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[1348] Examples:

[1349] Examples of data users submit:

[1350] User ID: student123

[1351] Query: When is the next math test?

[1352] 2. Query Parsing and Logging

[1353] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[1354] 3. Emotional Recognition

[1355] The server analyzes the user's emotions from the query using an emotion engine, which uses natural language processing techniques to identify the emotions contained in the query content.

[1356] Examples:

[1357] Emotions recognized by the emotion engine: anxiety, tension, anger, etc.

[1358] Query:

[1359] User ID: student123

[1360] Query: When is the next math test?

[1361] Recognized emotion: Anxiety

[1362] 4. Generating the Response

[1363] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[1364] Examples:

[1365] Response to the query:

[1366] The next math test is next Friday.

[1367] 5. Adjusting the response

[1368] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[1369] Examples:

[1370] Adjusted response:

[1371] "The next math test is next Friday. Don't worry."

[1372] 6. Returning a Response

[1373] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[1374] Examples:

[1375] The final response returned to the user:

[1376] "The next math test is next Friday. Don't worry."

[1377] Prompt Sentence Examples

[1378] User ID: user456

[1379] Query: Tell us about any recent security incidents.

[1380] Emotion: Anxiety

[1381] As described above, the system of the present invention is capable of providing appropriate responses based on the user's emotions 24 hours a day, thereby improving user satisfaction and reliability.

[1382] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1383] Step 1:

[1384] A user uses a device to send a query to an API endpoint. As input, the device sends JSON data containing the user ID and the query content. The device then sends this data as an HTTP POST request.

[1385] Specific behavior: The user enters a question into the application and presses the "Submit" button to send the data.

[1386] Step 2:

[1387] The server analyzes the request received. It receives a JSON-formatted query as input, extracts the user ID and the query content from it, and records this information and the current timestamp in a log file.

[1388] Specific behavior: The server parses the received data and logs it as follows. For example: "Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?"

[1389] Step 3:

[1390] The server uses an emotion engine to analyze the user's emotion from the query. Using the query content as input, the emotion engine identifies the emotion using natural language processing techniques. The identified emotion is obtained as output.

[1391] Specific behavior: The emotion engine identifies the emotion "anxiety" for the query "When is the next math test?"

[1392] Step 4:

[1393] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The input is the query and the analyzed emotion, and the AI ​​model generates a response based on the past query history and the emotion. The output is the generated response.

[1394] How it works: The AI ​​model generates the response "The next math test is next Friday" based on past data and the emotion "anxiety."

[1395] Step 5:

[1396] The server adjusts the generated response based on the user's emotion. The inputs are the generated response and the user's emotion, and the server takes into account the output of the emotion engine to adjust the response. The output is the adjusted response.

[1397] Behavior: Adjust the generated response "The next math test is next Friday" by adding the phrase "Don't worry, it's okay."

[1398] Step 6:

[1399] The server converts the adjusted response to JSON format and sends it to the user's device as an HTTP response. The input is the adjusted response, which the server converts to JSON format and sends to the user. The output is the response that is displayed on the user's device.

[1400] Specific behavior: In response to a question submitted by the user, the device displays a response saying, "The next math test is next Friday. Don't worry."

[1401] 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.

[1402] 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.

[1403] 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.

[1404] [Fourth embodiment]

[1405] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1406] 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.

[1407] 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).

[1408] 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.

[1409] 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.

[1410] 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).

[1411] 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.

[1412] 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.

[1413] 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.

[1414] 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.

[1415] 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.

[1416] 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.

[1417] 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."

[1418] The present invention is a system that uses AI models to automate customer support for inquiries from faculty, students, and other stakeholders at educational institutions. The system includes the following main functions:

[1419] 1. Receiving queries from users

[1420] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[1421] 2. Query Parsing and Logging

[1422] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[1423] 3. Generating a Response

[1424] The server generates appropriate responses to user queries using an AI model that is self-learning based on past query history and is constantly tuned to provide the best possible response.

[1425] 4. Returning a Response

[1426] The generated response is converted to JSON format and sent to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[1427] Natural language explanation of the process

[1428] Query reception

[1429] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[1430] json

[1431] {

[1432] "user_id": "student123",

[1433] "query": "When is the next math test?"

[1434] }

[1435] Query Parsing and Logging

[1436] The server receives this request and parses the user_id and query values, which are logged as follows:

[1437] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1438] Generating a response

[1439] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model is actually self-trained based on past query data, which improves the accuracy of the answer given to the user. An example response might be the following text:

[1440] The next math test is next Friday.

[1441] Returning a response

[1442] The server converts the generated response into JSON format and returns it to the user's device as an HTTP response, which the user sees on their device and sees something like this:

[1443] json

[1444] {

[1445] "response": "The next math test is next Friday."

[1446] }

[1447] The system operates in a 24-hour server environment, allowing users to make inquiries at any time, reducing the burden on educational institution staff and providing quick and accurate responses.

[1448] The processing flow will be explained below.

[1449] Step 1:

[1450] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[1451] json

[1452] {

[1453] "user_id": "student123",

[1454] "query": "When is the next math test?"

[1455] }

[1456] Step 2:

[1457] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[1458] Step 3:

[1459] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[1460] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1461] Step 4:

[1462] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. This function uses a self-trained AI model based on past query history to return the best possible response. In this simple example, the query is returned as the response.

[1463] Step 5:

[1464] The server converts the generated response into JSON format by using the jsonify function to package the generated text as a JSON object, for example generating a JSON response like this:

[1465] json

[1466] {

[1467] "response": "The next math test is next Friday."

[1468] }

[1469] Step 6:

[1470] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[1471] AI chatbot response: The next math test is next Friday.

[1472] The above processing steps enable the institution to provide quick and accurate responses to inquiries from faculty, students, and other stakeholders, thereby reducing staff workload and improving customer satisfaction.

[1473] Example 1

[1474] 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."

[1475] Educational institutions require fast and accurate responses to inquiries from teachers, students, and other stakeholders, but they face a shortage of human resources to handle these needs. Furthermore, the wide variety of inquiries means that individual responses are limited, making it particularly difficult to respond at night or on holidays. To solve this problem, automated systems are needed, but typical automated systems have low accuracy, making it difficult to improve user satisfaction.

[1476] 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.

[1477] In this invention, the server includes a means for a user to send a query to an API endpoint using a terminal, a means for the server to analyze the received query, extract the user ID and the query content, and record them in a log file, a means for the server to generate a response to the query based on past inquiry data using a generative AI model, and a means for returning the generated response to the user's terminal in JSON format, thereby reducing the burden on staff at the educational institution and enabling quick and accurate responses.

[1478] "User" refers to a teacher, student or other party of an educational institution who accesses the system and submits a query.

[1479] "Device" refers to electronic devices used by users, such as smartphones, PCs, and tablets.

[1480] "API endpoint" refers to a server-side interface through which users can submit queries.

[1481] A "query" refers to a question or inquiry that a user poses to a system.

[1482] "Server" refers to a central device for receiving queries from users, analyzing them, and generating responses.

[1483] "User ID" refers to a unique identifier used to identify a user on the system.

[1484] "JSON format" refers to a lightweight text-based data exchange format for structuring and sending data.

[1485] A "log file" is a file that records the system's operation history and events.

[1486] A "generative AI model" refers to an artificial intelligence model that self-learns based on past inquiry data and generates appropriate responses.

[1487] "Past inquiry data" refers to the history of queries and responses the system has received in the past.

[1488] "HTTP response" refers to a communication method used by a server to return a response generated in response to a user's query.

[1489] The present invention is a system for automating inquiries from teachers, students, and other stakeholders at educational institutions. The system accepts queries from users, analyzes the queries, generates appropriate responses using a generative AI model, and returns the responses in real time. A specific implementation of this system is described below.

[1490] First, a user uses a device such as a smartphone or PC to send a query to an API endpoint. This query is sent as an HTTP POST request, and the data is included in JSON format. As an example, consider the case where a user sends the following query:

[1491] json

[1492] {

[1493] "user_id": "student123",

[1494] "query": "When is the next math test?"

[1495] }

[1496] The server then receives this HTTP POST request, parses the JSON data to extract the user ID and the query, and records this information in a log file along with the current timestamp, thus preserving the complete query history for future analysis and troubleshooting.

[1497] The server uses a generative AI model to generate an appropriate response to the received query. This AI model self-learns based on past query data, so it can always provide the best response. For example, for the following query:

[1498] "Example of a prompt to input to a generative AI model: When is the next math test?"

[1499] The generative AI model generates a response like this:

[1500] The next math test is next Friday.

[1501] Finally, the server converts the generated response into JSON format and sends it back to the user's device as an HTTP response, where the user can view it and obtain the required information.

[1502] This system operates in a server environment that operates 24 hours a day, so users can make inquiries at any time and receive prompt and accurate answers. Another major advantage is that it significantly reduces the workload of educational institution staff.

[1503] The above is a specific embodiment of the present invention.

[1504] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1505] Step 1:

[1506] A user uses a terminal to send a query to an API endpoint. As input, they provide a JSON data containing the user ID and the question. For example, they send a query like this:

[1507] json

[1508] {

[1509] "user_id": "student123",

[1510] "query": "When is the next math test?"

[1511] }

[1512] This data is sent from the device to the server over the Internet as an HTTP POST request. The output is the request data that arrives at the server.

[1513] Step 2:

[1514] The server parses the JSON data of the received HTTP POST request. The input contains the received JSON data. Specifically, it extracts the user_id and query from the request body.

[1515] The output is the parsed data (user ID and query content). The specific data processing at this stage involves parsing the JSON data and extracting fields.

[1516] Step 3:

[1517] The server records the extracted user ID and query content along with the current timestamp in a log file. The input contains the parsed data (user ID, query content, timestamp). The specific operation is to format this information and write it to a log file.

[1518] The output is a successful logging event. This process also persists data and maintains history.

[1519] Step 4:

[1520] The server calls the function generate_response() to generate an appropriate response to the input query. The input includes the parsed query content. The generative AI model self-trains based on past inquiry data to generate highly accurate answers.

[1521] For example, the following text is output in response to the query "When is the next math test?":

[1522] The next math test is next Friday.

[1523] Specifically, the operation involves inputting a query into an AI model and generating a response.

[1524] Step 5:

[1525] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response. The input includes the response text from the generative AI model. The specific operation is to encode the response text into JSON format and send it to the user's device using a communication protocol.

[1526] The output is the response data that can be viewed on the user's terminal. This process involves formatting and sending the data.

[1527] Step 6:

[1528] The user sees the response from the server on the device. The input includes the response data sent from the server in JSON format. The specific operation is that the device parses the response and displays it to the user in a human-readable format.

[1529] The output is a response that the user can see, allowing the user to get a quick and accurate answer.

[1530] (Application example 1)

[1531] 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."

[1532] Responding to inquiries at educational institutions and brick-and-mortar stores requires a great deal of effort and manpower. Responding to inquiries outside of business hours is particularly difficult at brick-and-mortar stores, posing a risk of damaging customer satisfaction. Furthermore, the wide variety of inquiries makes it difficult to provide consistent responses, increasing the burden on employees. In these circumstances, there is a need for an efficient and accurate automated response system.

[1533] 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.

[1534] In this invention, the server includes means for generating responses to user queries using an AI model, means for recording received user queries and user IDs, means for generating responses to the received queries and returning the responses in JSON format, and means for automating responses to customer inquiries in physical stores. This enables effective and efficient response to inquiries in educational institutions and physical stores. Furthermore, a server environment that can operate 24 hours a day allows responses even outside of business hours, improving customer satisfaction.

[1535] An "AI model" is a mathematical model that uses artificial intelligence to accomplish a specific task.

[1536] A "query" is an inquiry made by a user to a system, and refers to a specific question or request.

[1537] A "response" is an answer that an AI model generates based on a query, and refers to the information it provides in response to a user's question.

[1538] A "user ID" is a unique identifier for identifying a specific user and is used to manage user information.

[1539] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for structuring and representing data.

[1540] A "brick and mortar store" is a commercial establishment that offers goods or services in a physical location.

[1541] "Automation" refers to the process of manually performing a task automatically using machines or software.

[1542] A "server environment" is a collection of hardware and software that provides the computing resources and network required for a system to operate.

[1543] A "log" is data that a system uses to record various events and transactions, and is stored for later reference.

[1544] "Self-learning" refers to the learning process by which an AI model improves its performance based on past data.

[1545] This invention shows how to realize an automatic response system for customer inquiries in a physical store. Specifically, the system is constructed as follows using a server, a user terminal, and an AI model.

[1546] System configuration

[1547] Hardware and Software

[1548] Server: A server environment that can operate 24 hours a day (e.g., AWS EC2 server) that runs the following software:

[1549] Server software: Python and Flask are used to provide API endpoints.

[1550] AI Model: A generative AI model built using TensorFlow or PyTorch.

[1551] Log management: Use a log management library such as loguru.

[1552] User Device: A smartphone (iOS or Android) used by a customer that runs the following software:

[1553] Mobile App: An application developed using React Native.

[1554] System Operation Overview

[1555] Query reception

[1556] A user enters a query using a mobile app on their smartphone and sends it as an HTTP POST request in JSON format to the server's API endpoint. This request includes the user ID and the query content.

[1557] Query Parsing and Logging

[1558] The server analyzes the user ID and query content from the received request and records this information in a log file along with the current timestamp, thus saving the query history.

[1559] Generating a response

[1560] The server calls the AI ​​model to generate an appropriate response to the query. The AI ​​model is self-trained based on past query data and is trained to provide a response appropriate to the query. The generated response is converted into JSON format.

[1561] Returning a response

[1562] The server returns the generated response to the user's device as an HTTP response, and the user can check the answer in real time on the device's application.

[1563] Specific examples

[1564] As an example, the processing flow when a customer sends a query such as "What are the opening hours of this store?" is shown below.

[1565] User submits the query:

[1566] json

[1567] {

[1568] "user_id": "customer456",

[1569] "query": "What are the opening hours for this store?"

[1570] }

[1571] Example of what the server logs:

[1572] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[1573] Response generation by AI model:

[1574] This store is open from 9am to 8pm.

[1575] Response that the server returns to the user:

[1576] json

[1577] {

[1578] "response": "This store is open from 9:00 AM to 8:00 PM."

[1579] }

[1580] Prompt Sentence Examples

[1581] Examples of specific prompts include:

[1582] Generate a response for the query: "What are this store's business hours?" based on the historical data.

[1583] This will enable stores to respond to customer inquiries quickly and accurately. In addition, a server environment that can operate 24 hours a day means customers can be served even outside of business hours, improving customer satisfaction.

[1584] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1585] Step 1:

[1586] A user inputs a query using a terminal.

[1587] Input: User ID and query content

[1588] Specific operation: The user opens the app on their smartphone, enters a question such as "What are the opening hours of this store?" and presses the send button.

[1589] Output: HTTP POST request in JSON format

[1590] json

[1591] {

[1592] "user_id": "customer456",

[1593] "query": "What are the opening hours for this store?"

[1594] }

[1595] Step 2:

[1596] The server receives the query and performs the analysis.

[1597] Input: HTTP POST request in JSON format

[1598] Specific operation: The server receives a request at an API endpoint built with Flask. It extracts the user ID and query content from the request.

[1599] Output: User ID and query content

[1600] User ID: customer456

[1601] Query: What are the opening hours for this store?

[1602] Step 3:

[1603] The server records parsed query information in a log file.

[1604] Input: User ID, query content, and timestamp

[1605] Specific operation: The server uses the Loguru library to record the received information in a log file.

[1606] Output: Recorded in a log file

[1607] Timestamp: 2023-10-02 16:00:00, User ID: customer456, Query: What are the opening hours of this store?

[1608] Step 4:

[1609] The server invokes the AI ​​model to generate a response to the query.

[1610] Input: Query content

[1611] How it works: The server loads a TensorFlow or PyTorch model and inputs the query content into the AI ​​model, which generates a response based on the trained data.

[1612] Output: The generated response

[1613] This store is open from 9am to 8pm.

[1614] Step 5:

[1615] The server converts the generated response into JSON format and sends it back to the user device.

[1616] Input: Generated response

[1617] What happens: The server uses Python's JSON library to serialize the response into JSON format, which it then sends back as an HTTP response.

[1618] Output: HTTP response in JSON format

[1619] json

[1620] {

[1621] "response": "This store is open from 9:00 AM to 8:00 PM."

[1622] }

[1623] Step 6:

[1624] The user terminal receives and displays the response from the server.

[1625] Input: HTTP response in JSON format

[1626] Specific behavior: The user device receives the HTTP response, deserializes it appropriately within the app, and displays it. The user confirms the answer displayed: "This store's business hours are from 9:00 AM to 8:00 PM."

[1627] Output: The response displayed on the screen

[1628] This store is open from 9am to 8pm.

[1629] 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.

[1630] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[1631] This system has the following main functions:

[1632] 1. Receiving queries from users

[1633] A user sends a query to an API endpoint using a device (such as a smartphone or PC). The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[1634] 2. Query Parsing and Logging

[1635] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[1636] 3. Emotional Recognition

[1637] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, if the query is "When is the next math test?", emotions such as tension and anxiety will be recognized.

[1638] 4. Generating the Response

[1639] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[1640] 5. Adjusting the response

[1641] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[1642] 6. Returning a Response

[1643] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[1644] Natural language explanation of the process

[1645] Query reception

[1646] The user sends a query to the API endpoint using their device, which is received by the server. The data sent is in JSON format, and contains, for example, the following:

[1647] json

[1648] {

[1649] "user_id": "student123",

[1650] "query": "When is the next math test?"

[1651] }

[1652] Query Parsing and Logging

[1653] The server receives this request and parses the user_id and query values, which are logged as follows:

[1654] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1655] Emotion recognition

[1656] The server uses an emotion engine to analyze the user's emotions from the query, for example, recognizing emotions such as "anxiety" or "tension" from the query.

[1657] Generating a response

[1658] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model, which uses the past query history and the output of the emotion engine to generate the response.

[1659] Response adjustment

[1660] The server tailors the generated response based on the user's emotions, for example, choosing reassuring words if the user is feeling anxious.

[1661] Returning a response

[1662] The server converts the response to JSON format and returns it to the user's device as an HTTP response, which the user sees on their device, looking something like this:

[1663] json

[1664] {

[1665] "response": "The next math test is next Friday. Don't worry."

[1666] }

[1667] The system allows faculty, students and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[1668] The processing flow will be explained below.

[1669] Step 1:

[1670] A user uses a device to send a query to an API endpoint. The data sent is an HTTP POST request in JSON format, containing the user ID and the query content. For example, it looks like this:

[1671] json

[1672] {

[1673] "user_id": "student123",

[1674] "query": "When is the next math test?"

[1675] }

[1676] Step 2:

[1677] The server receives a request at the API endpoint / api / query. The Flask framework handles this request and extracts the submitted data from request.json. Specifically, it extracts the user_id and query values.

[1678] Step 3:

[1679] The server gets the current time and logs the user_id, query, and timestamp to a log file, which stores the query history. For example, the following is logged:

[1680] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1681] Step 4:

[1682] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing algorithms to identify the emotion contained in the query text. For example, it can recognize emotions such as "anxiety" or "tension" from the query "When is the next math test?"

[1683] Step 5:

[1684] The server calls the function generate_response() to generate an appropriate response to the query using the AI ​​model. The AI ​​model takes the query history and the output of the emotion engine as input and generates an optimal response. Here, the response is generated taking emotions into account.

[1685] Step 6:

[1686] The server adjusts the generated response based on the user's emotions. For example, if the user is feeling anxious, it adds reassuring words to the response text. Specifically, it generates a response of the form, "The next math test is next Friday. Don't worry."

[1687] Step 7:

[1688] The server then converts this response to JSON by using the jsonify function to package the resulting text as a JSON object, resulting in a JSON response like this:

[1689] json

[1690] {

[1691] "response": "The next math test is next Friday. Don't worry."

[1692] }

[1693] Step 8:

[1694] The server sends this JSON response as an HTTP response to the user's device. The user can receive and display this response on their device. For example, the following appears on the device screen:

[1695] AI chatbot replies: The next math test is next Friday. Don't worry.

[1696] The above processing steps enable the provision of fast and accurate responses to inquiries from faculty, students, and other stakeholders of educational institutions. Responses that take emotion into consideration can provide users with more satisfactory service.

[1697] Example 2

[1698] 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."

[1699] Educational institutions need to respond quickly and appropriately to inquiries from faculty, students, and other stakeholders, but doing so manually is labor-intensive and time-consuming, and the quality of responses varies. Furthermore, there is a need for systems that can respond appropriately to queries that contain emotions, but conventional systems have difficulty recognizing user emotions and generating appropriate responses based on those emotions.

[1700] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1701] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording the received user query and user ID, means for analyzing the received query to recognize the user's emotion, means for adjusting the generated response based on the recognized emotion, and means for returning the generated response in JSON format, thereby enabling the generation and return of a prompt and appropriate response that takes the user's emotion into consideration.

[1702] An "AI model" is a program that uses artificial intelligence algorithms to recognize patterns and generate responses.

[1703] A "user" is an individual or entity that uses the system to make an inquiry.

[1704] A "query" is a question or request that a user sends to the system.

[1705] "User ID" means a unique identifier used by the system to identify a user.

[1706] An "emotion engine" is a software component that uses natural language processing to determine a user's emotion from text.

[1707] "JSON format" is an abbreviation for JavaScript Object Notation, and is a lightweight data exchange format for representing data in the form of JavaScript objects.

[1708] An "HTTP POST request" is a type of HTTP method for sending data to a web server.

[1709] A "log file" is a file that records system operations and user activities.

[1710] "Response adjustment" is the process of appropriately modifying the generated response based on the user's emotions.

[1711] An "HTTP response" is a response message that a server sends to a client.

[1712] This invention provides a customer support system that combines an AI model and an emotion engine to respond to inquiries from faculty, students, and other stakeholders in educational institutions. The system recognizes emotions and generates responses to user queries, and can respond 24 hours a day.

[1713] The system uses the following main hardware and software:

[1714] Hardware: Smartphones, PCs, servers

[1715] Software: AI model, emotion engine, logging system

[1716] The roles of each piece of hardware and software are as follows:

[1717] 1. Receiving queries

[1718] A user sends a query to the API endpoint using a device such as a smartphone or PC. This query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[1719] 2. Query Parsing and Logging

[1720] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[1721] 3. Emotional Recognition

[1722] The server uses an emotion engine to analyze the user's emotion from the query. This emotion engine uses natural language processing technology to identify the emotion contained in the query. For example, it can recognize emotions such as "anxiety" and "tension" from the query "When is the next math test?"

[1723] 4. Generating the Response

[1724] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[1725] 5. Adjusting the response

[1726] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[1727] 6. Returning a Response

[1728] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[1729] Examples of prompt statements

[1730] Here are some examples of prompts to input to a generative AI model:

[1731] Prompt statement example 1:

[1732] "When is the next math test?"

[1733] An example response to this:

[1734] "The next math test is next Friday. Don't worry."

[1735] Prompt statement example 2:

[1736] "Please tell me how to create a report card."

[1737] An example response to this:

[1738] "I will teach you how to create a report card. First, please log in to your school's portal site. Then, select the 'Report Card' tab and enter the required information. Please feel free to let me know if you have any additional questions."

[1739] This system allows faculty, students, and other stakeholders at educational institutions to receive prompt and appropriate responses, reducing staff workload and improving customer satisfaction.

[1740] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1741] Step 1: Query Reception

[1742] A user sends a query to the API endpoint using a smartphone or PC. The data sent is an HTTP POST request in JSON format. For example, the following input data is sent:

[1743] json

[1744] {

[1745] "user_id": "student123",

[1746] "query": "When is the next math test?"

[1747] }

[1748] The query is accepted when the device sends this JSON data to the API endpoint. The input includes the user ID and the query content. The output is received by the server.

[1749] Step 2: Parsing and logging queries

[1750] The server parses the incoming HTTP POST request and extracts the user_id and query values. For example, given the following input data:

[1751] json

[1752] {

[1753] "user_id": "student123",

[1754] "query": "When is the next math test?"

[1755] }

[1756] The server records the extracted user ID "student123" and the query content "When is the next math test?" along with a timestamp in a log file. Data processing here involves parsing the received data and generating log data.

[1757] As output, this will produce a log entry like this:

[1758] Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?

[1759] Step 3: Recognize emotions

[1760] The server inputs the query content into the emotion engine to recognize the user's emotion. The input in this process is the query "When is the next math test?"

[1761] The server analyzes the query with its emotion engine and recognizes the emotion, for example, "anxiety." As output, it gets the following data:

[1762] json

[1763] {

[1764] "emotion": "anxiety"

[1765] }

[1766] Specific operations include an emotion engine evaluating the query content using natural language processing algorithms.

[1767] Step 4: Generate a response

[1768] The server uses an AI model to generate a response based on the query and the perceived emotion. In this process, the query "When is the next math test?" and the emotion "anxiety" are input.

[1769] The server uses the function generate_response() to generate a response like this:

[1770] json

[1771] {

[1772] "response": "The next math test is next Friday. Don't worry."

[1773] }

[1774] The data calculation here is the generation of response text by an AI model.

[1775] Step 5: Adjust the response

[1776] The server adjusts the generated response based on the recognized emotion. Since the emotion is "anxious," the response includes a reassuring expression. The inputs are the generated response and the recognized emotion.

[1777] For example, the original response, "The next math test is next Friday," is adjusted to "The next math test is next Friday. Don't worry." The output is the adjusted response.

[1778] Step 6: Returning the response

[1779] The server then converts the final adjusted response into JSON format and sends it to the user's device as an HTTP response, with the adjusted response text as input.

[1780] json

[1781] {

[1782] "response": "The next math test is next Friday. Don't worry."

[1783] }

[1784] The server sends this JSON data to the user's device, where they can view it in real time. The output is an HTTP response containing the response.

[1785] (Application example 2)

[1786] 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."

[1787] Conventional customer support systems have the problem of being unable to properly grasp user emotions, resulting in inappropriate responses and a decline in user satisfaction. Furthermore, since it is difficult to provide 24-hour support, delays in response can lead to a loss of user trust. In the security field, in particular, it is necessary to quickly alleviate user anxiety and tension, so these issues must be resolved.

[1788] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1789] In this invention, the server includes means for generating a response to a query from a user using an AI model, means for recording received user queries and user IDs, means for generating a response to the received query and returning the response in JSON format, means for analyzing the user's emotion from the query using an emotion engine, and means for adjusting the generated response based on the user's emotion, thereby making it possible to provide appropriate responses according to the user's emotion 24 hours a day.

[1790] "AI model" refers to artificial intelligence that generates appropriate responses to user queries.

[1791] A "query" refers to a question or request that a user makes to a system.

[1792] "User ID" means a unique identifier used to identify a User.

[1793] "JSON format" stands for JavaScript Object Notation and refers to a way of representing data in a lightweight text format.

[1794] An "emotion engine" refers to software or a system for analyzing emotions from user queries.

[1795] "Self-learning" refers to the process of learning from past inquiry history to improve the accuracy and adaptability of responses.

[1796] "Server environment" refers to the hardware and software infrastructure on which the system operates.

[1797] The system of the present invention generates appropriate responses to user queries, adjusts the responses based on the user's emotions, and provides them on a 24-hour basis. A specific embodiment of the system is described below.

[1798] The system mainly consists of a server, a user terminal and related software.

[1799] System Configuration

[1800] 1. Server

[1801] The server is a computer with a powerful CPU and sufficient memory that receives, processes, and transmits data. Software used includes Python 3, Flask (or Django), etc. The server has the following features:

[1802] AI Model: Used to generate appropriate responses to user queries.

[1803] Emotion Engine: Uses natural language processing techniques to analyze user emotions from queries.

[1804] 2. User Device

[1805] A user device is a smartphone, computer, or other device used to send questions or requests. Users use their device to send queries to API endpoints.

[1806] Processing flow

[1807] 1. Receiving queries from users

[1808] A user uses a device to send a query to an API endpoint. The query is sent as an HTTP POST request in JSON format, and includes the user ID and the query content.

[1809] Examples:

[1810] Examples of data users submit:

[1811] User ID: student123

[1812] Query: When is the next math test?

[1813] 2. Query Parsing and Logging

[1814] The server parses the incoming request, extracts the user ID and the query, and records this information in a log file along with the current timestamp.

[1815] 3. Emotional Recognition

[1816] The server analyzes the user's emotions from the query using an emotion engine, which uses natural language processing techniques to identify the emotions contained in the query content.

[1817] Examples:

[1818] Emotions recognized by the emotion engine: anxiety, tension, anger, etc.

[1819] Query:

[1820] User ID: student123

[1821] Query: When is the next math test?

[1822] Recognized emotion: Anxiety

[1823] 4. Generating the Response

[1824] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The AI ​​model is self-trained based on past query history and also takes into account the output of the emotion engine to provide an appropriate response.

[1825] Examples:

[1826] Response to the query:

[1827] The next math test is next Friday.

[1828] 5. Adjusting the response

[1829] The generated response is tailored based on the output of the emotion engine: for example, if the user's emotion is anxious, the server will select a more kind and reassuring response.

[1830] Examples:

[1831] Adjusted response:

[1832] "The next math test is next Friday. Don't worry."

[1833] 6. Returning a Response

[1834] The server converts the generated response into JSON format and sends it to the user's device as an HTTP response, allowing the user to receive the answer in real time.

[1835] Examples:

[1836] The final response returned to the user:

[1837] "The next math test is next Friday. Don't worry."

[1838] Prompt Sentence Examples

[1839] User ID: user456

[1840] Query: Tell us about any recent security incidents.

[1841] Emotion: Anxiety

[1842] As described above, the system of the present invention is capable of providing appropriate responses based on the user's emotions 24 hours a day, thereby improving user satisfaction and reliability.

[1843] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1844] Step 1:

[1845] A user uses a device to send a query to an API endpoint. As input, the device sends JSON data containing the user ID and the query content. The device then sends this data as an HTTP POST request.

[1846] Specific behavior: The user enters a question into the application and presses the "Submit" button to send the data.

[1847] Step 2:

[1848] The server analyzes the request received. It receives a JSON-formatted query as input, extracts the user ID and the query content from it, and records this information and the current timestamp in a log file.

[1849] Specific behavior: The server parses the received data and logs it as follows. For example: "Timestamp: 2023-10-02 14:00:00, User ID: student123, Query: When is the next math test?"

[1850] Step 3:

[1851] The server uses an emotion engine to analyze the user's emotion from the query. Using the query content as input, the emotion engine identifies the emotion using natural language processing techniques. The identified emotion is obtained as output.

[1852] Specific behavior: The emotion engine identifies the emotion "anxiety" for the query "When is the next math test?"

[1853] Step 4:

[1854] The server uses an AI model to generate an appropriate response based on the user's query and the perceived emotion. The input is the query and the analyzed emotion, and the AI ​​model generates a response based on the past query history and the emotion. The output is the generated response.

[1855] How it works: The AI ​​model generates the response "The next math test is next Friday" based on past data and the emotion "anxiety."

[1856] Step 5:

[1857] The server adjusts the generated response based on the user's emotion. The inputs are the generated response and the user's emotion, and the server takes into account the output of the emotion engine to adjust the response. The output is the adjusted response.

[1858] Behavior: Adjust the generated response "The next math test is next Friday" by adding the phrase "Don't worry, it's okay."

[1859] Step 6:

[1860] The server converts the adjusted response to JSON format and sends it to the user's device as an HTTP response. The input is the adjusted response, which the server converts to JSON format and sends to the user. The output is the response that is displayed on the user's device.

[1861] Specific behavior: In response to a question submitted by the user, the device displays a response saying, "The next math test is next Friday. Don't worry."

[1862] 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.

[1863] 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.

[1864] 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 robot 414.

[1865] 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.

[1866] FIG. 9 illustrates 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 behaviors 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.

[1867] 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.

[1868] 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).

[1869] 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.

[1870] 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."

[1871] 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.

[1872] 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).

[1873] 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.

[1874] 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.

[1875] 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.

[1876] 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.

[1877] 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.

[1878] 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.

[1879] 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.

[1880] 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.

[1881] 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.

[1882] 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.

[1883] The following is further disclosed regarding the above embodiment.

[1884] (Claim 1)

[1885] means for generating responses to user queries using the AI ​​model;

[1886] means for recording received user queries and user IDs;

[1887] a means for generating a response to the received query and returning the response in JSON format;

[1888] A system including:

[1889] (Claim 2)

[1890] 10. The system of claim 1, further comprising means for self-learning using past inquiry history for response generation.

[1891] (Claim 3)

[1892] 10. The system of claim 1, further comprising means for operating in a 24-hour server environment.

[1893] "Example 1"

[1894] (Claim 1)

[1895] A means for a user to use a terminal to send queries to an API endpoint;

[1896] A means for analyzing the query received by the server, extracting the user ID and the query content, and recording them in a log file;

[1897] a means for the server to generate a response to the query based on past inquiry data using the generative AI model;

[1898] A means of returning the generated response to the user's device in JSON format;

[1899] A system including:

[1900] (Claim 2)

[1901] 10. The system of claim 1, wherein the system uses a generative AI model that is self-learning using past query history.

[1902] (Claim 3)

[1903] 10. The system of claim 1, which operates in a server environment that is available 24 hours a day.

[1904] "Application Example 1"

[1905] (Claim 1)

[1906] means for generating responses to user queries using the AI ​​model;

[1907] means for recording received user queries and user IDs;

[1908] a means for generating a response to the received query and returning the response in JSON format;

[1909] A means for automating responses to customer inquiries in a physical store;

[1910] A system including:

[1911] (Claim 2)

[1912] 10. The system of claim 1, further comprising means for self-learning using past inquiry history for response generation.

[1913] (Claim 3)

[1914] 10. The system of claim 1, further comprising means for operating in a 24-hour server environment.

[1915] "Example 2: Combining Emotion Engines"

[1916] (Claim 1)

[1917] means for generating responses to user queries using the AI ​​model;

[1918] means for recording received user queries and user IDs;

[1919] means for analyzing the received query to recognize a user's sentiment;

[1920] means for adjusting the generated response based on the recognized emotion;

[1921] A means to return the generated response in JSON format;

[1922] A system including:

[1923] (Claim 2)

[1924] 10. The system of claim 1, further comprising means for self-learning using past inquiry history for response generation.

[1925] (Claim 3)

[1926] 10. The system of claim 1, further comprising means for operating in a 24-hour server environment.

[1927] "Application example 2 when combining emotion engines"

[1928] (Claim 1)

[1929] means for generating responses to user queries using the AI ​​model;

[1930] means for recording received user queries and user IDs;

[1931] a means for generating a response to the received query and returning the response in JSON format;

[1932] means for parsing user sentiment from the query using a sentiment engine;

[1933] means for adjusting the generated response based on the user's emotions;

[1934] A system including:

[1935] (Claim 2)

[1936] 10. The system of claim 1, further comprising means for self-learning using past inquiry history for response generation.

[1937] (Claim 3)

[1938] 10. The system of claim 1, further comprising means for operating in a 24-hour server environment. [Explanation of symbols]

[1939] 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 generating responses to user queries using the AI ​​model; means for recording received user queries and user IDs; a means for generating a response to the received query and returning the response in JSON format; A system including:

2. The system of claim 1 further comprising means for self-learning using past inquiry history for response generation.

3. 10. The system of claim 1, further comprising means for operating in a server environment that is available 24 hours a day.

Images