system

Abstract

We provide personalized support to users and a system that improves the accuracy of support through data accumulation and learning. [Solution] The system includes a means for a user to input the consultation content in text format, a means for receiving the consultation content in a server and performing an initial analysis using natural language processing, a means for classifying the consultation content into topics based on the results of the initial analysis, a means for querying a database of past consultation content and extracting similar consultation content and its solutions, and a means for organizing the extracted solutions and generating proposals to be provided to the user. The system includes: a means for transmitting the generated suggestion content to the user's terminal and displaying it to the user; a means for receiving feedback from the user and storing the feedback in a database; and a means for training a machine learning model based on the stored data to generate more accurate support for the next consultation.

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] Providing an environment where users with mental health problems can feel safe and comfortable seeking advice is always a challenge. Another problem is that users who want to avoid face-to-face consultations have a high barrier to receiving appropriate support. Furthermore, there are concerns about confidentiality and a sense that simply asking for advice through an automated response system is not enough. Existing systems lack the ability to provide personalized support to users or to improve the accuracy of support through data accumulation and learning. [Means for solving the problem]

[0005] The above problem is solved by providing a system including the following means:

[0006] A means for users to input their consultation details in text format;

[0007] a means for receiving the consultation content in a server and performing an initial analysis using natural language processing;

[0008] A means for classifying the consultation contents into topics based on the initial analysis results;

[0009] A means for searching a database of past consultation contents and extracting similar consultation contents and their solutions;

[0010] a means for organizing the extracted solutions and generating proposals to be provided to a user;

[0011] means for transmitting the generated proposal content to a user's terminal and displaying it to the user;

[0012] means for receiving feedback from users and storing the feedback in a database;

[0013] A means for training a machine learning model based on the accumulated data to generate more accurate support for the next consultation;

[0014] The system is provided with:

[0015] This system allows users to receive professional support anonymously and with peace of mind, avoiding face-to-face consultations, and provides personalized suggestions tailored to the user's situation.In addition, by accumulating feedback in a database and training a learning model, the system's support accuracy can be continuously improved.

[0016] "User" refers to an individual who accesses the system, inputs their inquiry details, and receives proposals.

[0017] "Terminal" refers to electronic devices such as personal computers and smartphones used by users.

[0018] "Server" refers to a computer system that receives the user's inquiry content, analyzes it, and generates proposal content.

[0019] "Consultation content" refers to textual information about worries or problems entered by the user.

[0020] "Natural language processing" refers to technology for analyzing text data and understanding human language.

[0021] "Initial analysis" refers to the first stage of data analysis carried out after receiving the consultation content.

[0022] "Sentiment analysis" refers to the technique of assessing the emotional tone of text data.

[0023] "Topic classification" refers to the process of classifying input consultation content into specific categories.

[0024] A "database" refers to a collection of data that stores past consultation details and their solutions and makes them searchable.

[0025] "Similar consultation content" refers to cases where it is determined that there is a certain degree of similarity between the current consultation content and the consultation content included in the database of past consultation content.

[0026] "Solutions" refer to measures or advice that are deemed to be effective based on past consultations.

[0027] "Recommendation" refers to advice or solution information provided to the User that is generated based on analysis and database queries.

[0028] "Feedback" refers to the impressions and evaluations that users enter regarding the suggestions provided by the system.

[0029] A "machine learning model" refers to an algorithm that learns from accumulated data and improves the accuracy of support from the next time onwards.

[0030] "Emergency Support" refers to immediate assistance provided when a user's issue is deemed to be of an extremely serious nature.

[0031] "Alert" means a warning or notification issued when it is determined that emergency support is required. [Brief explanation of the drawings]

[0032] [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

[0033] 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.

[0034] First, the terms used in the following description will be explained.

[0035] 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).

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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."

[0040] [First embodiment]

[0041] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0042] 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.

[0043] 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).

[0044] 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.

[0045] 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.

[0046] 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.

[0047] 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.

[0048] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0049] 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.

[0050] 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.

[0051] 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.

[0052] 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."

[0053] This invention provides a system that allows users to receive mental health consultations and appropriate support without having to deal with face-to-face consultations. The system is composed of multiple means. Each means and its specific processing will be described below.

[0054] User Access and Login

[0055] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[0056] Enter and submit your inquiry

[0057] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[0058] Receiving consultation content and initial analysis

[0059] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, it performs sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[0060] Historical database queries

[0061] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[0062] Proposal generation and delivery

[0063] The server sends the generated suggestions to the device, which displays them in the chat interface and allows the user to review the advice or solution provided.

[0064] Receive feedback and learn

[0065] The user inputs feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation.

[0066] Specific examples

[0067] Example 1: Work stress

[0068] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[0069] The terminal transmits the consultation contents to the server.

[0070] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigning a low emotional score and classifying the topic as "work stress."

[0071] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[0072] The server generates the proposal and sends it to the device.

[0073] The device displays the suggestions to the user.

[0074] Users try out the proposed methods and provide feedback.

[0075] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[0076] Example 2: Relationship problems

[0077] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[0078] The terminal transmits the consultation contents to the server.

[0079] The server analyzes text data containing the keywords "colleague" and "poor relationship" and classifies sentiment scores and topics.

[0080] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[0081] The server generates the proposal and sends it to the device.

[0082] The device displays the suggestions to the user.

[0083] The user reviews the solution and provides feedback.

[0084] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[0085] This system provides users with an environment where they can seek advice anonymously and with peace of mind, and continuous feedback improves the accuracy of the system.

[0086] The processing flow will be explained below.

[0087] Step 1:

[0088] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[0089] Step 2:

[0090] The device displays a login screen and the user enters their username and password.

[0091] Step 3:

[0092] The terminal transmits the entered authentication information to the server.

[0093] Step 4:

[0094] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[0095] Step 5:

[0096] The server displays a dashboard or chat interface to the authenticated user.

[0097] Step 6:

[0098] The user enters the content of the consultation in text format into the chat interface.

[0099] Step 7:

[0100] The terminal transmits the input consultation content to the server.

[0101] Step 8:

[0102] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[0103] Step 9:

[0104] As part of the initial analysis, the server performs sentiment analysis to calculate a sentiment score for the consultation content, and also performs topic classification to assign the content to the appropriate category.

[0105] Step 10:

[0106] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[0107] Step 11:

[0108] The server organizes the extracted solutions and generates proposals.

[0109] Step 12:

[0110] The server sends the generated proposal to the terminal.

[0111] Step 13:

[0112] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[0113] Step 14:

[0114] The user enters feedback on the provided suggestions into a chat interface.

[0115] Step 15:

[0116] The device sends the user's feedback to the server.

[0117] Step 16:

[0118] The server receives the feedback and stores it in a database.

[0119] Step 17:

[0120] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[0121] Step 18:

[0122] The next time the user uses the system, the server will use the updated machine learning model to provide more personalized support to the user.

[0123] Example 1

[0124] 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."

[0125] Conventional mental health consultation systems have difficulty providing appropriate support to users who wish to avoid in-person consultations. Furthermore, these systems lack sufficient mechanisms for effectively utilizing user feedback to improve the accuracy of subsequent consultations. Furthermore, they often lack the means to properly analyze the content of the consultation and provide appropriate solutions. To address these issues, the present invention provides a system that allows users to safely seek mental health consultations and receive appropriate support while avoiding in-person consultations.

[0126] 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.

[0127] In this invention, the server includes: means for a user to input a consultation content in text format; means for a terminal to transmit the consultation content to the server; means for the server to receive the consultation content and perform an initial analysis using natural language processing technology; means for classifying the consultation content by topic based on the results of the initial analysis; means for querying a database of past consultation content and extracting similar consultation content and solutions; means for organizing the extracted solutions and generating a proposal content to be provided to the user; means for transmitting the generated proposal content to the user's terminal and displaying it to the user; means for receiving feedback from the user and storing the feedback in a database; and means for training a machine learning model based on the stored data to generate more accurate support for the next consultation. This allows users to receive mental health consultation anonymously and with peace of mind, and by utilizing continuous feedback, the system can provide more accurate support for subsequent consultations.

[0128] "User" refers to a person who uses the system to seek mental health advice.

[0129] "Terminal" refers to a device, such as a personal computer or smartphone, that a user uses to access the system and enter consultation details.

[0130] "Server" refers to a computer system that processes the inquiry content sent by the user and generates and provides proposal content.

[0131] "Consultation content" refers to text information entered by users about questions and concerns about mental health.

[0132] "Natural language processing technology" refers to technology that analyzes text data, calculates sentiment scores, and performs topic classification.

[0133] "Initial analysis" refers to the initial analysis process that the server performs after receiving the consultation content using natural language processing technology.

[0134] "Topic classification" refers to the process of categorizing consultation content into specific categories or themes based on the results of the initial analysis.

[0135] A "database" refers to a collection of information that stores and manages past consultation details and their solutions.

[0136] A "machine learning model" refers to an algorithm or system that is trained to provide more accurate support for the next consultation based on feedback accumulated in a database.

[0137] "Sentiment analysis" refers to the technology of calculating emotional trends and emotional scores from text data of consultation content.

[0138] "Emotion score" refers to the numerical representation of the intensity and tendency of emotions contained in the consultation content as a result of emotion analysis.

[0139] "Proposal" refers to advice or solutions generated by the server and provided to the user.

[0140] "Feedback" refers to the ratings and comments that users enter in response to suggestions.

[0141] This invention relates to a system that allows users to receive mental health consultations and appropriate support in a safe environment while avoiding face-to-face consultations. The specific configuration and operation of this system will be described below.

[0142] User Access and Login

[0143] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The entered authentication information is sent from the device to the server. The server verifies the authentication information, and if authentication is successful, starts a session and displays a dashboard and chat interface to the user.

[0144] Enter and submit your inquiry

[0145] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[0146] Receiving consultation content and initial analysis

[0147] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology (e.g., Google® Cloud Natural Language API, AWS® Comprehend, etc.). The analysis includes sentiment analysis and topic classification. Sentiment analysis calculates an emotional score for the consultation content, and topic classification classifies the consultation content into specific categories.

[0148] Historical database queries

[0149] Based on the results of the initial analysis, the server queries a database of past consultations (e.g., MySQL (registered trademark), PostgreSQL, etc.) to extract similar consultations and their solutions. The server then organizes these extracted solutions and generates proposals to provide to users.

[0150] Proposal generation and delivery

[0151] The server generates suggestions and sends them to the device, which displays them in a chat interface, allowing the user to review the advice and solutions provided.

[0152] Receive feedback and learn

[0153] The user enters feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores the feedback in a database and trains a machine learning model to provide more accurate support for the next consultation.

[0154] Specific examples

[0155] Example 1: Work stress

[0156] 1. A user says, "I haven't been able to sleep lately because of work stress."

[0157] 2. The device sends the consultation details to the server.

[0158] 3. The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigns a low emotional score, and classifies the topic as "work stress."

[0159] 4. The server queries the historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[0160] 5. The server generates the proposal and sends it to the device.

[0161] 6. The device displays the suggestions to the user.

[0162] 7. The user tries out the proposed solution and provides feedback.

[0163] 8. The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[0164] Example 2: Relationship problems

[0165] 1. A user comes to us with a problem: "I'm having trouble with my coworkers."

[0166] 2. The device sends the consultation details to the server.

[0167] 3. The server analyzes text data containing the keywords "colleague" and "relationship not going well" and classifies sentiment scores and topics.

[0168] 4. The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[0169] 5. The server generates the proposal and sends it to the device.

[0170] 6. The device displays the suggestions to the user.

[0171] 7. The user reviews the solution and provides feedback.

[0172] 8. The device sends the feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[0173] Prompt Sentence Examples

[0174] "I haven't been able to sleep lately because of work stress. What should I do?"

[0175] "I'm having trouble with my coworkers. How can I improve it?"

[0176] In this way, users can seek mental health advice anonymously and with peace of mind, and by utilizing continuous feedback, the system will be able to provide more accurate support for subsequent consultations.

[0177] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0178] Step 1:

[0179] A user accesses the system's website or application using a device such as a personal computer or smartphone. The input is a username and password, and the device displays a login screen, which collects the authentication information. The output is the authentication information entered by the user.

[0180] Step 2:

[0181] The device sends the entered authentication information to the server. The input is the username and password entered by the user, and the output is the authentication information sent to the server. The device securely transmits the authentication information using the SSL / TLS protocol.

[0182] Step 3:

[0183] The server queries the user database based on the received authentication information to verify authentication. The input is the received authentication information, and the output is whether the authentication was successful. If authentication is successful, the server generates a session ID and generates information to display the dashboard and chat interface.

[0184] Step 4:

[0185] The user inputs the content of the consultation in text format into the chat interface. The input is the text of the consultation content, and the output is the input text data.

[0186] Step 5:

[0187] The terminal transmits the input consultation content to the server. The input is text data entered by the user, and the output is text data transmitted to the server. The terminal transmits the data using a communication protocol.

[0188] Step 6:

[0189] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. The input is the received text data, and the output is the analysis results. The server performs sentiment analysis, calculates sentiment scores, and performs topic classification.

[0190] Step 7:

[0191] The server queries a database of past consultations based on the initial analysis results. The input is the initial analysis results, and the output is similar consultations and their solutions. The server queries the database and organizes the extracted solutions.

[0192] Step 8:

[0193] The server generates a proposal to be provided to the user based on the extracted solution. The input is the extracted solution, and the output is the proposal. The server sends the generated proposal to the terminal.

[0194] Step 9:

[0195] The terminal displays the suggestions in a chat interface, and the user confirms the offered advice or solution. The input is the suggestion sent from the server, and the output is the suggestion displayed in the chat interface.

[0196] Step 10:

[0197] The user inputs feedback on the provided suggestion. The input is the feedback, and the output is the input feedback.

[0198] Step 11:

[0199] The terminal sends the user's feedback to the server. The input is the feedback content, and the output is the feedback sent to the server. The terminal transmits the data using a communication protocol.

[0200] Step 12:

[0201] The server accumulates the feedback in a database and trains a machine learning model. The input is the feedback content, and the output is the trained machine learning model. This makes it possible to provide more accurate support for the next consultation.

[0202] (Application example 1)

[0203] 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."

[0204] Mental health care for drivers in autonomous vehicles is an important issue for ensuring safe driving and reducing driver stress. However, conventional systems have no way for drivers to receive appropriate support in real time while avoiding face-to-face consultations. This has led to a lack of mental health follow-up in emergencies while driving.

[0205] 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.

[0206] In this invention, the server includes means for a user to input a consultation content in text format, means for receiving the consultation content in the server and performing an initial analysis using natural language processing, means for topic categorizing the consultation content based on the results of the initial analysis, means for consulting a database of past consultation content and extracting similar consultation contents and their solutions, means for organizing the extracted solutions and generating a proposal content to be provided to the user, means for sending the generated proposal content to the user's terminal and displaying it to the user, means for receiving feedback from the user and storing the feedback in a database, means for training a machine learning model based on the stored data and generating more accurate support for the next consultation, means for providing an interface for providing mental health care in an autonomous vehicle, and means for monitoring the driver's mental state while driving and providing support in real time, thereby enabling the driver to receive mental health care in real time in an autonomous vehicle.

[0207] "User" refers to a driver who uses the system to seek mental health advice.

[0208] "Text consultation content" refers to the string of characters that users enter into the system, and the content represents mental health issues and concerns.

[0209] The "server" is the computer at the core of the system, which receives inquiries from users and analyzes them and provides support suggestions.

[0210] "Natural language processing" is a technology that allows a computer to analyze text data entered by a user and understand its meaning and emotions.

[0211] "Initial analysis" is the process of first analyzing the consultation content using natural language processing technology and extracting emotion scores and keywords.

[0212] "Topic classification" refers to classifying the content of a consultation into multiple categories based on the results of the initial analysis.

[0213] The "past consultation content database" is a database that stores consultation content and solutions that have been accumulated in the system up to now.

[0214] "Similar consultation content" refers to cases in the database of past consultation content that are similar in content and emotion to the current consultation content.

[0215] "Solutions" are advice or measures provided to resolve a user's mental health issues based on past consultations.

[0216] The "proposal content" is a support proposal generated for the current consultation content based on the extracted solution.

[0217] "User's device" means a device used by a user, including a smartphone, tablet, or in-vehicle interface.

[0218] "Feedback" refers to the opinions and evaluations that users give in response to suggestions, and is used to improve the accuracy of the system.

[0219] A "machine learning model" is an algorithm that learns from large amounts of data and generates more accurate support for the next consultation.

[0220] An "autonomous vehicle" is a vehicle that automatically performs driving operations and provides a place for the driver to receive mental health care.

[0221] An "interface" is an operating means, such as a screen or voice input system, that allows a user to interact with a system.

[0222] "Mental state monitoring" is the process of monitoring a driver's emotions and stress levels and collecting data in real time.

[0223] This invention is a system that allows users to receive mental health consultations and appropriate support while in an autonomous vehicle. The specific configuration and procedure are shown below.

[0224] 1. System Configuration

[0225] Main components:

[0226] User devices: devices such as touchscreens in vehicles and smartphones

[0227] Server: Cloud-based computing cluster

[0228] Natural language processing models, such as BERT for Hugging Face

[0229] Database: A database for storing past consultations and feedback

[0230] 2. User Access and Login

[0231] Users access the system using a touchscreen in their vehicle or a smartphone, enter their username and password, and the device sends the credentials to a server. After authentication, the session begins and the user is presented with a dashboard and chat interface.

[0232] 3. Enter and submit your inquiry

[0233] The user inputs the text of the consultation into the chat interface, and the device sends the inputted consultation content to the server.

[0234] 4. Receiving consultation content and initial analysis

[0235] When the server receives the consultation content, it performs an initial analysis of the text data using a natural language processing model. As part of the analysis, it performs a sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[0236] 5. Query the database and generate suggestions

[0237] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions, and then organizes the extracted solutions to generate proposals to be provided to the user.

[0238] 6. Providing suggestions and feedback

[0239] The server generates suggestions and sends them to the user's device, which displays them in the chat interface. The user reviews the advice and solutions provided and enters feedback on the results. The device then sends the feedback to the server, which stores it in a database.

[0240] 7. Feedback learning and model training

[0241] The server uses the accumulated feedback to train a machine learning model and provide more accurate support for the next consultation.

[0242] 8. Operation in autonomous vehicles

[0243] The system can monitor the driver's mental state while driving in an autonomous vehicle and provide appropriate support in real time. For example, if the driver feels anxious on the highway, the system can suggest deep breathing or meditation to relieve tension.

[0244] Specific examples

[0245] 1. Example 1: Highway anxiety

[0246] Consultation: "Recently, I've been feeling anxious when driving on the highway."

[0247] Suggestion: "To reduce highway anxiety, try listening to appropriate music and taking deep breaths."

[0248] Feedback: "I tried the deep breathing exercises you suggested and my anxiety has decreased a little."

[0249] 2. Example 2: Commuting stress

[0250] Consultation: "I always feel stressed during my morning commute."

[0251] Suggestion: "Trying a short meditation session before you start driving may help reduce stress."

[0252] Feedback: "I tried meditation and it made me feel a little less stressed in the mornings."

[0253] In this way, the driver can receive real-time mental health care in the vehicle, providing a safe and comfortable driving environment.

[0254] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0255] Step 1:

[0256] The user enters their username and password using the vehicle's touchscreen or smartphone.

[0257] Input: Username, Password

[0258] Output: Credentials

[0259] Specific operation:

[0260] The user operates the login screen on the vehicle's touchscreen and enters their username and password.

[0261] Step 2:

[0262] The terminal transmits the entered authentication information to the server, and the server performs authentication.

[0263] Input: Credentials

[0264] Output: Authentication result (success / failure)

[0265] Specific operation:

[0266] The device sends the user's input information to the server via encrypted communication such as SSL, and the server compares it with a database and returns the authentication result.

[0267] Step 3:

[0268] If authentication is successful, the server starts a session and presents the user with a dashboard and chat interface.

[0269] Input: Authentication result (success)

[0270] Output: Dashboard, chat interface

[0271] Specific operation:

[0272] The server generates a session ID and sends it to the user's device, which receives it and displays the dashboard and chat interface.

[0273] Step 4:

[0274] The user inputs the content of the consultation in text format into the chat interface, and the terminal transmits the content to the server.

[0275] Input: Consultation content (text format)

[0276] Output: Sending the consultation details to the server

[0277] Specific operation:

[0278] A user types "Lately, I feel anxious when I drive on the highway" into the chat interface and clicks the send button. The device then sends this text to the server.

[0279] Step 5:

[0280] The server receives the consultation content and performs an initial analysis using a natural language processing model.

[0281] Input: Consultation content (text format)

[0282] Output: Initial analysis results (sentiment score, topic classification)

[0283] Specific operation:

[0284] The server performs sentiment analysis using Hugging Face's BERT model and other tools to analyze the content of the consultation. The analysis results include "anxiety (NEGATIVE)."

[0285] Step 6:

[0286] The server queries a database of past consultation contents based on the results of the initial analysis and extracts similar consultation contents and their solutions.

[0287] Input: Initial analysis results (sentiment score, topic classification)

[0288] Output: Similar inquiries and solutions

[0289] Specific operation:

[0290] The server uses the analysis results to query a database and extract similar consultations and past solutions (e.g., deep breathing, meditation, etc.).

[0291] Step 7:

[0292] The server organizes the extracted solutions and generates proposals to be presented to the user.

[0293] Input: Similar inquiries and solutions

[0294] Output: Proposal

[0295] Specific operation:

[0296] The server organizes solutions and generates suggestions such as, "To reduce anxiety on the highway, try listening to appropriate music or taking deep breaths."

[0297] Step 8:

[0298] The server sends the generated proposals to the user's device, which then displays them to the user.

[0299] Input: Proposal

[0300] Output: what is displayed to the user

[0301] Specific operation:

[0302] The server transmits the proposal to the terminal, which displays the proposal on the chat interface.

[0303] Step 9:

[0304] The user tries the proposed method and inputs feedback on the results, which is then sent to the server.

[0305] Input: Feedback (text format)

[0306] Output: Send feedback

[0307] Specific operation:

[0308] The user types, "I tried the suggested deep breathing exercises and my anxiety has decreased a little," and submits the feedback. The device then sends this feedback to the server.

[0309] Step 10:

[0310] The server receives the feedback, stores it in a database, and trains a machine learning model to provide more accurate support for the next consultation.

[0311] Input: Feedback

[0312] Output: A trained machine learning model

[0313] Specific operation:

[0314] The server stores the feedback in a database and uses this data to update and train the machine learning model, improving the accuracy of support for the next consultation.

[0315] 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.

[0316] The present invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing are explained below.

[0317] User Access and Login

[0318] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[0319] Enter and submit your inquiry

[0320] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[0321] Receiving consultation content and initial analysis

[0322] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The server also categorizes the consultation content into topics and assigns them to appropriate categories.

[0323] Historical database queries

[0324] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[0325] Emotional engine regulation

[0326] The server uses an emotion engine to dynamically adjust its suggestions based on the user's emotion score, providing advice and solutions that are optimized for the user's emotions.

[0327] Proposal generation and delivery

[0328] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[0329] Receive feedback and learn

[0330] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[0331] Specific examples

[0332] Example 1: Work stress

[0333] A user complains, "I haven't been able to sleep lately because of work stress."

[0334] The terminal transmits the consultation contents to the server.

[0335] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[0336] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[0337] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[0338] The server sends the proposal to the device.

[0339] The device displays the suggestions to the user.

[0340] Users try out the proposed methods and provide feedback.

[0341] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[0342] Example 2: Relationship problems

[0343] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[0344] The terminal transmits the consultation contents to the server.

[0345] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[0346] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[0347] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[0348] The server sends the proposal to the device.

[0349] The device displays the suggestions to the user.

[0350] The user implements the proposed method and provides feedback.

[0351] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[0352] This system provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, it provides personalized support optimized to the user's emotions.Continuous feedback constantly improves the system's accuracy, resulting in more effective support.

[0353] The processing flow will be explained below.

[0354] Step 1:

[0355] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[0356] Step 2:

[0357] The device displays a login screen and the user enters their username and password.

[0358] Step 3:

[0359] The terminal transmits the entered authentication information to the server.

[0360] Step 4:

[0361] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[0362] Step 5:

[0363] The server displays a dashboard or chat interface to the authenticated user.

[0364] Step 6:

[0365] The user enters the content of the consultation in text format into the chat interface.

[0366] Step 7:

[0367] The terminal transmits the input consultation content to the server.

[0368] Step 8:

[0369] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[0370] Step 9:

[0371] As part of the initial analysis, the server uses an emotion engine to calculate an emotion score and recognize the emotion of the consultation content.

[0372] Step 10:

[0373] The server classifies the consultation topic and assigns it to the appropriate category.

[0374] Step 11:

[0375] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[0376] Step 12:

[0377] The server organizes the extracted solutions and dynamically adjusts the proposals based on the emotion score using an emotion engine.

[0378] Step 13:

[0379] The server sends the generated proposal to the terminal.

[0380] Step 14:

[0381] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[0382] Step 15:

[0383] The user enters feedback on the provided suggestions into a chat interface.

[0384] Step 16:

[0385] The device sends the user's feedback to the server.

[0386] Step 17:

[0387] The server receives the feedback and stores it in a database.

[0388] Step 18:

[0389] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[0390] Step 19:

[0391] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[0392] Step 20:

[0393] The next time the system is used, the server will use the updated machine learning model and emotional data to provide more accurate, optimized support to the user.

[0394] Example 2

[0395] 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."

[0396] In recent years, mental health issues have been increasing, creating a demand for effective and prompt support. However, it is difficult to provide an anonymous environment where users can seek advice in a safe and secure environment. Furthermore, there are limited systems that provide appropriate advice based on the user's emotions and the content of their consultation. As a result, the accuracy and effectiveness of responses to users' consultations are often insufficient.

[0397] 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.

[0398] In this invention, the server includes a means for a user to input the consultation content in text format, a means for receiving the consultation content in the server and performing an initial analysis using natural language processing, and a means for dynamically adjusting the content of suggestions based on the user's emotion score. This provides an environment where users can consult anonymously with peace of mind, and by using an emotion engine, it becomes possible to provide personalized support optimized for the user's emotions in real time.

[0399] "User" refers to an individual who accesses the system and inputs the details of their consultation.

[0400] "Terminal" refers to a device used by a user, such as a personal computer or smartphone.

[0401] "Server" refers to a central computer system that receives and analyzes consultation content, generates proposals, and provides them to users.

[0402] "Natural language processing" refers to the technology for analyzing input text data and understanding its meaning and emotions.

[0403] "Initial analysis" refers to the process of first analyzing the text of the input consultation content, calculating the emotion score, and classifying topics.

[0404] An "emotion engine" refers to an algorithm or model that calculates an emotional score from text data entered by the user and recognizes their emotional state.

[0405] "Emotion score" refers to a numerical value that indicates the user's emotional state, calculated based on the input text data.

[0406] "Topic classification" refers to the process of sorting the input consultation content into appropriate categories based on its content.

[0407] "Database" refers to a storage device or system for storing past consultation details and feedback data.

[0408] A "machine learning model" refers to an algorithm or program that is trained using collected data to provide more accurate support for the next consultation.

[0409] "Suggestion" refers to advice or solutions generated by the server and provided to the user.

[0410] "Feedback" refers to the opinions and evaluations that users enter regarding the suggestions.

[0411] "Dynamic adjustment" refers to changing the suggestions in real time in response to changes in the user's emotional score, etc.

[0412] "Emergency Support" refers to fast, special support provided when a user's inquiry requires urgent attention.

[0413] "Alert" means a warning or notification generated when it is determined that emergency support is required.

[0414] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system is characterized by incorporating an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing will be explained below.

[0415] User Access and Login

[0416] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[0417] Enter and submit your inquiry

[0418] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[0419] Receiving consultation content and initial analysis

[0420] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The consultation content is also classified into topics and sorted into appropriate categories. Natural language processing technology used includes Python's NLTK and SpaCy.

[0421] Historical database queries

[0422] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions. It then organizes the extracted solutions and generates proposals to be provided to users. To achieve this, a database management system (e.g., MySQL or PostgreSQL) is used.

[0423] Emotional engine regulation

[0424] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score, providing advice and solutions optimized for the user's emotions. The emotion engine uses emotion analysis models such as BERT and RoBERTa.

[0425] Proposal generation and delivery

[0426] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[0427] Receive feedback and learn

[0428] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[0429] Specific examples

[0430] Example 1: Work stress

[0431] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[0432] The terminal transmits the consultation contents to the server.

[0433] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[0434] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[0435] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[0436] The server sends the proposal to the device.

[0437] The device displays the suggestions to the user.

[0438] Users try out the proposed methods and provide feedback.

[0439] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[0440] Example 2: Relationship problems

[0441] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[0442] The terminal transmits the consultation contents to the server.

[0443] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[0444] The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[0445] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[0446] The server sends the proposal to the device.

[0447] The device displays the suggestions to the user.

[0448] The user implements the proposed method and provides feedback.

[0449] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[0450] This invention provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, personalized support optimized for the user's emotions is provided in real time. Continuous feedback constantly improves the accuracy and effectiveness of the system.

[0451] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0452] Step 1:

[0453] A user accesses the system's website or application using a terminal such as a personal computer or smartphone.

[0454] Input: Website URL or application icon

[0455] Output: Login screen displayed

[0456] What happens: The user enters a URL in a web browser or taps an application to launch it. The device renders and displays a login screen to the user.

[0457] Step 2:

[0458] The device will display a login screen and prompt you to enter your username and password.

[0459] Input: None

[0460] Output: Username and password input fields displayed

[0461] What it does: The terminal renders a login form with username and password input fields.

[0462] Step 3:

[0463] The user enters their username and password and clicks the submit button.

[0464] Input: Username and Password

[0465] Output: Send authentication information to the server

[0466] How it works: The user enters information on the keyboard and clicks the login button. The device sends the information to the server.

[0467] Step 4:

[0468] Verify the authentication information received by the server.

[0469] Input: Username and Password

[0470] Output: Authentication result

[0471] What it does: The server queries its database to verify that the username and password submitted match.

[0472] Step 5:

[0473] If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[0474] Input: Authentication result (success)

[0475] Output: Display URL for dashboard or chat interface

[0476] How it works: The server generates a session ID, saves the session data, and responds with a URL for the dashboard or chat interface, which the device receives and displays to the user.

[0477] Step 6:

[0478] The user enters the content of the consultation in text format into the chat interface.

[0479] Input: Text of consultation

[0480] Output: Consultation content data

[0481] Action: The user enters text using a keyboard or touchscreen and clicks the send button. The device sends the entered text data to the server.

[0482] Step 7:

[0483] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[0484] Input: Consultation content data

[0485] Output: Initial analysis results (tokenized text and sentiment scores)

[0486] How it works: The server uses Python libraries such as NLTK and SpaCy to tokenize and morphologically analyze the text and calculate a sentiment score.

[0487] Step 8:

[0488] The server categorizes the consultation content into topics based on the initial analysis results.

[0489] Input: Initial analysis results

[0490] Output: Topic classification results

[0491] How it works: The server uses topic models (LDA or LSI) to classify text data into categories.

[0492] Step 9:

[0493] The server queries a database of past consultations and extracts similar consultations and their solutions.

[0494] Input: Topic classification results

[0495] Output: Similar issues and solutions

[0496] How it works: The server generates an SQL query and sends it to the database to retrieve similar queries and their solutions.

[0497] Step 10:

[0498] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score.

[0499] Input: Similar issues and solutions, sentiment scores

[0500] Output: Dynamically adjusted suggestions

[0501] How it works: The server adjusts the tone and content of the solution it presents based on the sentiment score.

[0502] Step 11:

[0503] The server sends the generated proposal to the terminal.

[0504] Input: Dynamically adjusted suggestions

[0505] Output: Proposal content data

[0506] How it works: The server sends the suggestions as an HTTP response, which the device receives and renders in the chat interface.

[0507] Step 12:

[0508] Users enter feedback on the suggestions provided into a chat interface.

[0509] Input: Feedback text

[0510] Output: Feedback data

[0511] Action: The user enters feedback in text format and clicks the send button. The device sends the feedback data to the server.

[0512] Step 13:

[0513] The server accumulates user feedback in a database and trains a machine learning model.

[0514] Input: Feedback data

[0515] Output: An updated machine learning model

[0516] How it works: The server stores the feedback data in a database and periodically retrains the machine learning model.

[0517] Step 14:

[0518] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[0519] Input: User's historical sentiment score

[0520] Output: Personalized support

[0521] How it works: The server analyzes historical data of sentiment scores and applies an algorithm to track long-term sentiment trends.

[0522] (Application example 2)

[0523] 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."

[0524] While existing mental health support systems provide an environment where users can seek advice anonymously and with peace of mind, they do not adequately respond appropriately to the user's emotions. Furthermore, they lacked mechanisms for effectively utilizing user feedback to improve the accuracy of future consultations, making it difficult to provide optimal support tailored to each user's individual situation. Furthermore, there were no systems that could suggest foods or beverages based on the content of mental health consultations, meaning there was a lack of a means to provide immediate, specific support for specific emotions or physical conditions.

[0525] 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.

[0526] In this invention, the server includes: a means for a user to input a text-based consultation content; a means for receiving the consultation content in the server and performing an initial analysis using natural language processing; a means for topic-classifying the consultation content based on the results of the initial analysis; a means for querying a database of past consultation content to extract similar consultation content and solutions; a means for organizing the extracted solutions and generating a proposal content to be provided to the user; a means for sending the generated proposal content to the user's terminal and displaying it to the user; a means for receiving feedback from the user and storing the feedback in a database; a means for training a machine learning model based on the accumulated data to generate more accurate support for the next consultation; a means for dynamically optimizing food and beverage suggestions based on the user's consultation content and emotion score; a means for generating food and beverage suggestions based on the user's emotion using a generative AI model; and a means for generating an alert if the generated proposal content determines that emergency support is required. This allows for the provision of proposal content optimized to the user's emotion, enabling immediate food and beverage suggestions tailored to specific emotions or physical conditions. Furthermore, by utilizing user feedback, the system's accuracy can be improved, enabling more personalized support to be provided.

[0527] "User" refers to an individual who uses the System.

[0528] "Text-format consultation content" refers to text information that a user inputs into the system, including consultations and questions.

[0529] "Server" refers to a computing system that receives, processes, and transmits data.

[0530] "Natural language processing" is a technology that uses computers to understand, analyze, and generate human language.

[0531] "Initial analysis" is the process of first analyzing the input text data to understand its basic structure and meaning.

[0532] "Topic classification" refers to the process of sorting text data into specific subjects or categories.

[0533] "Past consultation content database" refers to a data bank that stores consultation content and solutions that users have previously entered into the system.

[0534] "Similar consultation contents and their solutions" refers to consultation contents and their solutions that are highly similar to the current consultation contents and past consultation contents in the database.

[0535] "Suggestions" refers to the advice or solutions the system provides to the user.

[0536] "User Device" refers to the electronic device (e.g., smartphone, tablet, or PC) used by a User to access the System.

[0537] "Feedback" refers to the evaluations and opinions that users give regarding the suggestions provided.

[0538] A "machine learning model" refers to an algorithm that automatically learns from accumulated data and makes predictions and classifications.

[0539] "More precise support" refers to assistance that responds appropriately and effectively to the user's individual situation.

[0540] An "emotion score" is an index that numerically evaluates a user's emotional state based on their text input.

[0541] "Dynamically optimized food and drink" refers to food and drink suggestions that are tailored based on the user's emotions and physical condition.

[0542] "Generative AI model" refers to an artificial intelligence model used to perform a generative task.

[0543] "Emergency Support" means an emergency response required immediately by a User.

[0544] An "alert" is a notification that notifies you of an emergency or important information.

[0545] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. The system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions.

[0546] System configuration

[0547] This system includes a user device, a server, a database, a natural language processing engine, an emotion engine, and a machine learning model. The user device refers to an electronic device such as a smartphone, tablet, or PC, and the server is a computer system for receiving, processing, and transmitting data.

[0548] Program processing

[0549] Login and Access

[0550] A user accesses the system's website or application using their own device and enters their username and password to log in. This authentication information is sent to the server, and once the server has successfully authenticated, the user is provided with a chat interface for consultation.

[0551] Enter and submit your inquiry

[0552] The user enters the content of the consultation in text format into the chat interface, and this content is sent to the server.

[0553] Initial analysis and sentiment analysis

[0554] The server uses natural language processing technology to perform an initial analysis of the received consultation content, and then calculates an emotion score using an emotion engine, which allows the user's emotional state to be numerically evaluated.

[0555] Topic classification and historical data reference

[0556] Based on the analysis results, the consultation content is classified into topics, and similar consultation content and solutions are extracted by referring to a database of past consultation content.

[0557] Proposal generation and optimization

[0558] The server generates recommendations based on the extracted solutions and dynamically optimizes them based on the emotion score, suggesting foods and drinks that correspond to the user's emotions and physical condition.

[0559] View suggestions and get feedback

[0560] The generated suggestions are sent back to the user's device and displayed to the user. The user checks the suggestions and enters feedback. This feedback is sent to the server and stored in a database.

[0561] Training a machine learning model

[0562] The server uses the accumulated feedback to train a machine learning model and generate more accurate support for the next consultation.

[0563] Hardware and software used

[0564] Hardware: User devices (smartphones, tablets, PCs), servers (high-performance computing systems)

[0565] Software: Natural language processing engines (e.g., SpaCy, NLTK), emotion engines (e.g., Hugging Face transformers library), machine learning models (e.g., Scikit-learn, TENSORFLOW®)

[0566] Specific examples

[0567] Example 1:

[0568] If a user types, "I've been stressed out lately and have lost my appetite.":

[0569] The system suggests, "Would you like some relaxing herbal tea?"

[0570] The user orders the suggested herbal tea and enters feedback as "That was a good suggestion, I feel better."

[0571] Prompt Sentence Examples

[0572] "The user has recently been feeling stressed and has lost their appetite. Please suggest suitable foods and drinks. Use your historical database and emotion engine to generate the best suggestions."

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

[0574] Specific processing flow of the program

[0575] Step 1: User Login and Access

[0576] Input: The user enters their username and password from their device (smartphone, tablet, computer).

[0577] Data processing: The device sends the entered authentication information (username and password) to the server.

[0578] Data calculation: The server checks the authentication information against the database.

[0579] Output: If authentication is successful, the server starts a session and presents the user with a chat interface for consultation.

[0580] Step 2: Enter and submit your inquiry

[0581] Input: The user enters the content of their inquiry in text format into the chat interface.

[0582] Data processing: The device sends the consultation details entered to the server.

[0583] Data calculation: None (simple data transmission).

[0584] Output: The consultation content is received by the server.

[0585] Step 3: Initial analysis and sentiment analysis

[0586] Input: Text data of the consultation received by the server.

[0587] Data processing: The server performs initial analysis of the text data using a natural language processing engine (e.g., SpaCy, NLTK).

[0588] Data calculation: An emotion engine (e.g., Hugging Face transformers library) is used to calculate an emotion score and numerically evaluate the emotional state of the consultation.

[0589] Output: Analysis results of the text data and sentiment scores.

[0590] Step 4: Topic classification and historical data reference

[0591] Input: Initial analysis results and sentiment scores.

[0592] Data processing: The server categorizes the consultation content into topics based on the analysis results.

[0593] Data calculation: Query the database of past consultations to extract similar consultations and their solutions.

[0594] Output: Topic classification results, and similar consultations and their solutions.

[0595] Step 5: Generate and optimize proposals

[0596] Input: Similar consultations and their solutions, sentiment scores.

[0597] Data processing: The server generates suggestions using the generative AI model.

[0598] Data computation: Dynamically optimizes recommendations based on emotion scores, suggesting foods and drinks that correspond to the user's emotions and physical condition.

[0599] Output: Dynamically optimized suggestions.

[0600] Step 6: View your proposal and get feedback

[0601] Input: Dynamically optimized suggestions.

[0602] Data processing: The server sends the proposal to the user's device.

[0603] Data calculation: None (simple data transmission).

[0604] Output: The suggestion is displayed on the user's device.

[0605] Step 7: Capture and store feedback

[0606] Input: User feedback.

[0607] Data processing: The device sends feedback to the server.

[0608] Data calculation: The server stores the received feedback in a database.

[0609] Output: Feedback stored in a database.

[0610] Step 8: Train the machine learning model

[0611] Input: Accumulated feedback data.

[0612] Data processing: The server inputs the feedback data into the machine learning model.

[0613] Data Computing: Using machine learning models (e.g., Scikit-learn, TensorFlow) to train the next consultation to improve its accuracy.

[0614] Output: An improved machine learning model.

[0615] Specific examples

[0616] Example 1: If a user enters, "I've been feeling stressed lately and have no appetite," the system will suggest, "How about some herbal tea to help you relax?" and the user will confirm and enter their feedback.

[0617] Example prompt sentence:

[0618] "The user has recently been feeling stressed and has lost their appetite. Please suggest suitable foods and drinks. Use your historical database and emotion engine to generate the best suggestions."

[0619] 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.

[0620] 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.

[0621] 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.

[0622] [Second embodiment]

[0623] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0624] 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.

[0625] 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).

[0626] 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.

[0627] 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.

[0628] 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).

[0629] 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.

[0630] 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.

[0631] 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.

[0632] 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.

[0633] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0634] 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."

[0635] This invention provides a system that allows users to receive mental health consultations and appropriate support without having to deal with face-to-face consultations. The system is composed of multiple means. Each means and its specific processing will be described below.

[0636] User Access and Login

[0637] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[0638] Enter and submit your inquiry

[0639] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[0640] Receiving consultation content and initial analysis

[0641] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, it performs sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[0642] Historical database queries

[0643] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[0644] Proposal generation and delivery

[0645] The server sends the generated suggestions to the device, which displays them in the chat interface and allows the user to review the advice or solution provided.

[0646] Receive feedback and learn

[0647] The user inputs feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation.

[0648] Specific examples

[0649] Example 1: Work stress

[0650] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[0651] The terminal transmits the consultation contents to the server.

[0652] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigning a low emotional score and classifying the topic as "work stress."

[0653] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[0654] The server generates the proposal and sends it to the device.

[0655] The device displays the suggestions to the user.

[0656] Users try out the proposed methods and provide feedback.

[0657] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[0658] Example 2: Relationship problems

[0659] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[0660] The terminal transmits the consultation contents to the server.

[0661] The server analyzes text data containing the keywords "colleague" and "poor relationship" and classifies sentiment scores and topics.

[0662] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[0663] The server generates the proposal and sends it to the device.

[0664] The device displays the suggestions to the user.

[0665] The user reviews the solution and provides feedback.

[0666] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[0667] This system provides users with an environment where they can seek advice anonymously and with peace of mind, and continuous feedback improves the accuracy of the system.

[0668] The processing flow will be explained below.

[0669] Step 1:

[0670] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[0671] Step 2:

[0672] The device displays a login screen and the user enters their username and password.

[0673] Step 3:

[0674] The terminal transmits the entered authentication information to the server.

[0675] Step 4:

[0676] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[0677] Step 5:

[0678] The server displays a dashboard or chat interface to the authenticated user.

[0679] Step 6:

[0680] The user enters the content of the consultation in text format into the chat interface.

[0681] Step 7:

[0682] The terminal transmits the input consultation content to the server.

[0683] Step 8:

[0684] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[0685] Step 9:

[0686] As part of the initial analysis, the server performs sentiment analysis to calculate a sentiment score for the consultation content, and also performs topic classification to assign the content to the appropriate category.

[0687] Step 10:

[0688] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[0689] Step 11:

[0690] The server organizes the extracted solutions and generates proposals.

[0691] Step 12:

[0692] The server sends the generated proposal to the terminal.

[0693] Step 13:

[0694] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[0695] Step 14:

[0696] The user enters feedback on the provided suggestions into a chat interface.

[0697] Step 15:

[0698] The device sends the user's feedback to the server.

[0699] Step 16:

[0700] The server receives the feedback and stores it in a database.

[0701] Step 17:

[0702] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[0703] Step 18:

[0704] The next time the user uses the system, the server will use the updated machine learning model to provide more personalized support to the user.

[0705] Example 1

[0706] 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."

[0707] Conventional mental health consultation systems have difficulty providing appropriate support to users who wish to avoid in-person consultations. Furthermore, these systems lack sufficient mechanisms for effectively utilizing user feedback to improve the accuracy of subsequent consultations. Furthermore, they often lack the means to properly analyze the content of the consultation and provide appropriate solutions. To address these issues, the present invention provides a system that allows users to safely seek mental health consultations and receive appropriate support while avoiding in-person consultations.

[0708] 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.

[0709] In this invention, the server includes: means for a user to input a consultation content in text format; means for a terminal to transmit the consultation content to the server; means for the server to receive the consultation content and perform an initial analysis using natural language processing technology; means for classifying the consultation content by topic based on the results of the initial analysis; means for querying a database of past consultation content and extracting similar consultation content and solutions; means for organizing the extracted solutions and generating a proposal content to be provided to the user; means for transmitting the generated proposal content to the user's terminal and displaying it to the user; means for receiving feedback from the user and storing the feedback in a database; and means for training a machine learning model based on the stored data to generate more accurate support for the next consultation. This allows users to receive mental health consultation anonymously and with peace of mind, and by utilizing continuous feedback, the system can provide more accurate support for subsequent consultations.

[0710] "User" refers to a person who uses the system to seek mental health advice.

[0711] "Terminal" refers to a device, such as a personal computer or smartphone, that a user uses to access the system and enter consultation details.

[0712] "Server" refers to a computer system that processes the inquiry content sent by the user and generates and provides proposal content.

[0713] "Consultation content" refers to text information entered by users about questions and concerns about mental health.

[0714] "Natural language processing technology" refers to technology that analyzes text data, calculates sentiment scores, and performs topic classification.

[0715] "Initial analysis" refers to the initial analysis process that the server performs after receiving the consultation content using natural language processing technology.

[0716] "Topic classification" refers to the process of categorizing consultation content into specific categories or themes based on the results of the initial analysis.

[0717] A "database" refers to a collection of information that stores and manages past consultation details and their solutions.

[0718] A "machine learning model" refers to an algorithm or system that is trained to provide more accurate support for the next consultation based on feedback accumulated in a database.

[0719] "Sentiment analysis" refers to the technology of calculating emotional trends and emotional scores from text data of consultation content.

[0720] "Emotion score" refers to the numerical representation of the intensity and tendency of emotions contained in the consultation content as a result of emotion analysis.

[0721] "Proposal" refers to advice or solutions generated by the server and provided to the user.

[0722] "Feedback" refers to the ratings and comments that users enter in response to suggestions.

[0723] This invention relates to a system that allows users to receive mental health consultations and appropriate support in a safe environment while avoiding face-to-face consultations. The specific configuration and operation of this system will be described below.

[0724] User Access and Login

[0725] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The entered authentication information is sent from the device to the server. The server verifies the authentication information, and if authentication is successful, starts a session and displays a dashboard and chat interface to the user.

[0726] Enter and submit your inquiry

[0727] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[0728] Receiving consultation content and initial analysis

[0729] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology (e.g., Google Cloud Natural Language API, AWS Comprehend, etc.). The analysis includes sentiment analysis and topic classification. Sentiment analysis calculates an emotional score for the consultation content, and topic classification classifies the consultation content into specific categories.

[0730] Historical database queries

[0731] Based on the initial analysis results, the server queries a database of past consultations (e.g., MySQL, PostgreSQL, etc.) to extract similar consultations and their solutions. The server then organizes these extracted solutions and generates proposals to be provided to users.

[0732] Proposal generation and delivery

[0733] The server generates suggestions and sends them to the device, which displays them in a chat interface, allowing the user to review the advice and solutions provided.

[0734] Receive feedback and learn

[0735] The user enters feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores the feedback in a database and trains a machine learning model to provide more accurate support for the next consultation.

[0736] Specific examples

[0737] Example 1: Work stress

[0738] 1. A user says, "I haven't been able to sleep lately because of work stress."

[0739] 2. The device sends the consultation details to the server.

[0740] 3. The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigns a low emotional score, and classifies the topic as "work stress."

[0741] 4. The server queries the historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[0742] 5. The server generates the proposal and sends it to the device.

[0743] 6. The device displays the suggestions to the user.

[0744] 7. The user tries out the proposed solution and provides feedback.

[0745] 8. The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[0746] Example 2: Relationship problems

[0747] 1. A user comes to us with a problem: "I'm having trouble with my coworkers."

[0748] 2. The device sends the consultation details to the server.

[0749] 3. The server analyzes text data containing the keywords "colleague" and "relationship not going well" and classifies sentiment scores and topics.

[0750] 4. The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[0751] 5. The server generates the proposal and sends it to the device.

[0752] 6. The device displays the suggestions to the user.

[0753] 7. The user reviews the solution and provides feedback.

[0754] 8. The device sends the feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[0755] Prompt Sentence Examples

[0756] "I haven't been able to sleep lately because of work stress. What should I do?"

[0757] "I'm having trouble with my coworkers. How can I improve it?"

[0758] In this way, users can seek mental health advice anonymously and with peace of mind, and by utilizing continuous feedback, the system will be able to provide more accurate support for subsequent consultations.

[0759] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0760] Step 1:

[0761] A user accesses the system's website or application using a device such as a personal computer or smartphone. The input is a username and password, and the device displays a login screen, which collects the authentication information. The output is the authentication information entered by the user.

[0762] Step 2:

[0763] The device sends the entered authentication information to the server. The input is the username and password entered by the user, and the output is the authentication information sent to the server. The device securely transmits the authentication information using the SSL / TLS protocol.

[0764] Step 3:

[0765] The server queries the user database based on the received authentication information to verify authentication. The input is the received authentication information, and the output is whether the authentication was successful. If authentication is successful, the server generates a session ID and generates information to display the dashboard and chat interface.

[0766] Step 4:

[0767] The user inputs the content of the consultation in text format into the chat interface. The input is the text of the consultation content, and the output is the input text data.

[0768] Step 5:

[0769] The terminal transmits the input consultation content to the server. The input is text data entered by the user, and the output is text data transmitted to the server. The terminal transmits the data using a communication protocol.

[0770] Step 6:

[0771] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. The input is the received text data, and the output is the analysis results. The server performs sentiment analysis, calculates sentiment scores, and performs topic classification.

[0772] Step 7:

[0773] The server queries a database of past consultations based on the initial analysis results. The input is the initial analysis results, and the output is similar consultations and their solutions. The server queries the database and organizes the extracted solutions.

[0774] Step 8:

[0775] The server generates a proposal to be provided to the user based on the extracted solution. The input is the extracted solution, and the output is the proposal. The server sends the generated proposal to the terminal.

[0776] Step 9:

[0777] The terminal displays the suggestions in a chat interface, and the user confirms the offered advice or solution. The input is the suggestion sent from the server, and the output is the suggestion displayed in the chat interface.

[0778] Step 10:

[0779] The user inputs feedback on the provided suggestion. The input is the feedback, and the output is the input feedback.

[0780] Step 11:

[0781] The terminal sends the user's feedback to the server. The input is the feedback content, and the output is the feedback sent to the server. The terminal transmits the data using a communication protocol.

[0782] Step 12:

[0783] The server accumulates the feedback in a database and trains a machine learning model. The input is the feedback content, and the output is the trained machine learning model. This makes it possible to provide more accurate support for the next consultation.

[0784] (Application example 1)

[0785] 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."

[0786] Mental health care for drivers in autonomous vehicles is an important issue for ensuring safe driving and reducing driver stress. However, conventional systems have no way for drivers to receive appropriate support in real time while avoiding face-to-face consultations. This has led to a lack of mental health follow-up in emergencies while driving.

[0787] 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.

[0788] In this invention, the server includes means for a user to input a consultation content in text format, means for receiving the consultation content in the server and performing an initial analysis using natural language processing, means for topic categorizing the consultation content based on the results of the initial analysis, means for consulting a database of past consultation content and extracting similar consultation contents and their solutions, means for organizing the extracted solutions and generating a proposal content to be provided to the user, means for sending the generated proposal content to the user's terminal and displaying it to the user, means for receiving feedback from the user and storing the feedback in a database, means for training a machine learning model based on the stored data and generating more accurate support for the next consultation, means for providing an interface for providing mental health care in an autonomous vehicle, and means for monitoring the driver's mental state while driving and providing support in real time, thereby enabling the driver to receive mental health care in real time in an autonomous vehicle.

[0789] "User" refers to a driver who uses the system to seek mental health advice.

[0790] "Text consultation content" refers to the string of characters that users enter into the system, and the content represents mental health issues and concerns.

[0791] The "server" is the computer at the core of the system, which receives inquiries from users and analyzes them and provides support suggestions.

[0792] "Natural language processing" is a technology that allows a computer to analyze text data entered by a user and understand its meaning and emotions.

[0793] "Initial analysis" is the process of first analyzing the consultation content using natural language processing technology and extracting emotion scores and keywords.

[0794] "Topic classification" refers to classifying the content of a consultation into multiple categories based on the results of the initial analysis.

[0795] The "past consultation content database" is a database that stores consultation content and solutions that have been accumulated in the system up to now.

[0796] "Similar consultation content" refers to cases in the database of past consultation content that are similar in content and emotion to the current consultation content.

[0797] "Solutions" are advice or measures provided to resolve a user's mental health issues based on past consultations.

[0798] The "proposal content" is a support proposal generated for the current consultation content based on the extracted solution.

[0799] "User's device" means a device used by a user, including a smartphone, tablet, or in-vehicle interface.

[0800] "Feedback" refers to the opinions and evaluations that users give in response to suggestions, and is used to improve the accuracy of the system.

[0801] A "machine learning model" is an algorithm that learns from large amounts of data and generates more accurate support for the next consultation.

[0802] An "autonomous vehicle" is a vehicle that automatically performs driving operations and provides a place for the driver to receive mental health care.

[0803] An "interface" is an operating means, such as a screen or voice input system, that allows a user to interact with a system.

[0804] "Mental state monitoring" is the process of monitoring a driver's emotions and stress levels and collecting data in real time.

[0805] This invention is a system that allows users to receive mental health consultations and appropriate support while in an autonomous vehicle. The specific configuration and procedure are shown below.

[0806] 1. System Configuration

[0807] Main components:

[0808] User devices: devices such as touchscreens in vehicles and smartphones

[0809] Server: Cloud-based computing cluster

[0810] Natural language processing models, such as BERT for Hugging Face

[0811] Database: A database for storing past consultations and feedback

[0812] 2. User Access and Login

[0813] Users access the system using a touchscreen in their vehicle or a smartphone, enter their username and password, and the device sends the credentials to a server. After authentication, the session begins and the user is presented with a dashboard and chat interface.

[0814] 3. Enter and submit your inquiry

[0815] The user inputs the text of the consultation into the chat interface, and the device sends the inputted consultation content to the server.

[0816] 4. Receiving consultation content and initial analysis

[0817] When the server receives the consultation content, it performs an initial analysis of the text data using a natural language processing model. As part of the analysis, it performs a sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[0818] 5. Query the database and generate suggestions

[0819] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions, and then organizes the extracted solutions to generate proposals to be provided to the user.

[0820] 6. Providing suggestions and feedback

[0821] The server generates suggestions and sends them to the user's device, which displays them in the chat interface. The user reviews the advice and solutions provided and enters feedback on the results. The device then sends the feedback to the server, which stores it in a database.

[0822] 7. Feedback learning and model training

[0823] The server uses the accumulated feedback to train a machine learning model and provide more accurate support for the next consultation.

[0824] 8. Operation in autonomous vehicles

[0825] The system can monitor the driver's mental state while driving in an autonomous vehicle and provide appropriate support in real time. For example, if the driver feels anxious on the highway, the system can suggest deep breathing or meditation to relieve tension.

[0826] Specific examples

[0827] 1. Example 1: Highway anxiety

[0828] Consultation: "Recently, I've been feeling anxious when driving on the highway."

[0829] Suggestion: "To reduce highway anxiety, try listening to appropriate music and taking deep breaths."

[0830] Feedback: "I tried the deep breathing exercises you suggested and my anxiety has decreased a little."

[0831] 2. Example 2: Commuting stress

[0832] Consultation: "I always feel stressed during my morning commute."

[0833] Suggestion: "Trying a short meditation session before you start driving may help reduce stress."

[0834] Feedback: "I tried meditation and it made me feel a little less stressed in the mornings."

[0835] In this way, the driver can receive real-time mental health care in the vehicle, providing a safe and comfortable driving environment.

[0836] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0837] Step 1:

[0838] The user enters their username and password using the vehicle's touchscreen or smartphone.

[0839] Input: Username, Password

[0840] Output: Credentials

[0841] Specific operation:

[0842] The user operates the login screen on the vehicle's touchscreen and enters their username and password.

[0843] Step 2:

[0844] The terminal transmits the entered authentication information to the server, and the server performs authentication.

[0845] Input: Credentials

[0846] Output: Authentication result (success / failure)

[0847] Specific operation:

[0848] The device sends the user's input information to the server via encrypted communication such as SSL, and the server compares it with a database and returns the authentication result.

[0849] Step 3:

[0850] If authentication is successful, the server starts a session and presents the user with a dashboard and chat interface.

[0851] Input: Authentication result (success)

[0852] Output: Dashboard, chat interface

[0853] Specific operation:

[0854] The server generates a session ID and sends it to the user's device, which receives it and displays the dashboard and chat interface.

[0855] Step 4:

[0856] The user inputs the content of the consultation in text format into the chat interface, and the terminal transmits the content to the server.

[0857] Input: Consultation content (text format)

[0858] Output: Sending the consultation details to the server

[0859] Specific operation:

[0860] A user types "Lately, I feel anxious when I drive on the highway" into the chat interface and clicks the send button. The device then sends this text to the server.

[0861] Step 5:

[0862] The server receives the consultation content and performs an initial analysis using a natural language processing model.

[0863] Input: Consultation content (text format)

[0864] Output: Initial analysis results (sentiment score, topic classification)

[0865] Specific operation:

[0866] The server performs sentiment analysis using Hugging Face's BERT model and other tools to analyze the content of the consultation. The analysis results include "anxiety (NEGATIVE)."

[0867] Step 6:

[0868] The server queries a database of past consultation contents based on the results of the initial analysis and extracts similar consultation contents and their solutions.

[0869] Input: Initial analysis results (sentiment score, topic classification)

[0870] Output: Similar inquiries and solutions

[0871] Specific operation:

[0872] The server uses the analysis results to query a database and extract similar consultations and past solutions (e.g., deep breathing, meditation, etc.).

[0873] Step 7:

[0874] The server organizes the extracted solutions and generates proposals to be presented to the user.

[0875] Input: Similar inquiries and solutions

[0876] Output: Proposal

[0877] Specific operation:

[0878] The server organizes solutions and generates suggestions such as, "To reduce anxiety on the highway, try listening to appropriate music or taking deep breaths."

[0879] Step 8:

[0880] The server sends the generated proposals to the user's device, which then displays them to the user.

[0881] Input: Proposal

[0882] Output: what is displayed to the user

[0883] Specific operation:

[0884] The server transmits the proposal to the terminal, which displays the proposal on the chat interface.

[0885] Step 9:

[0886] The user tries the proposed method and inputs feedback on the results, which is then sent to the server.

[0887] Input: Feedback (text format)

[0888] Output: Send feedback

[0889] Specific operation:

[0890] The user types, "I tried the suggested deep breathing exercises and my anxiety has decreased a little," and submits the feedback. The device then sends this feedback to the server.

[0891] Step 10:

[0892] The server receives the feedback, stores it in a database, and trains a machine learning model to provide more accurate support for the next consultation.

[0893] Input: Feedback

[0894] Output: A trained machine learning model

[0895] Specific operation:

[0896] The server stores the feedback in a database and uses this data to update and train the machine learning model, improving the accuracy of support for the next consultation.

[0897] 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.

[0898] The present invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing are explained below.

[0899] User Access and Login

[0900] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[0901] Enter and submit your inquiry

[0902] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[0903] Receiving consultation content and initial analysis

[0904] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The server also categorizes the consultation content into topics and assigns them to appropriate categories.

[0905] Historical database queries

[0906] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[0907] Emotional engine regulation

[0908] The server uses an emotion engine to dynamically adjust its suggestions based on the user's emotion score, providing advice and solutions that are optimized for the user's emotions.

[0909] Proposal generation and delivery

[0910] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[0911] Receive feedback and learn

[0912] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[0913] Specific examples

[0914] Example 1: Work stress

[0915] A user complains, "I haven't been able to sleep lately because of work stress."

[0916] The terminal transmits the consultation contents to the server.

[0917] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[0918] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[0919] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[0920] The server sends the proposal to the device.

[0921] The device displays the suggestions to the user.

[0922] Users try out the proposed methods and provide feedback.

[0923] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[0924] Example 2: Relationship problems

[0925] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[0926] The terminal transmits the consultation contents to the server.

[0927] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[0928] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[0929] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[0930] The server sends the proposal to the device.

[0931] The device displays the suggestions to the user.

[0932] The user implements the proposed method and provides feedback.

[0933] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[0934] This system provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, it provides personalized support optimized to the user's emotions.Continuous feedback constantly improves the system's accuracy, resulting in more effective support.

[0935] The processing flow will be explained below.

[0936] Step 1:

[0937] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[0938] Step 2:

[0939] The device displays a login screen and the user enters their username and password.

[0940] Step 3:

[0941] The terminal transmits the entered authentication information to the server.

[0942] Step 4:

[0943] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[0944] Step 5:

[0945] The server displays a dashboard or chat interface to the authenticated user.

[0946] Step 6:

[0947] The user enters the content of the consultation in text format into the chat interface.

[0948] Step 7:

[0949] The terminal transmits the input consultation content to the server.

[0950] Step 8:

[0951] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[0952] Step 9:

[0953] As part of the initial analysis, the server uses an emotion engine to calculate an emotion score and recognize the emotion of the consultation content.

[0954] Step 10:

[0955] The server classifies the consultation topic and assigns it to the appropriate category.

[0956] Step 11:

[0957] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[0958] Step 12:

[0959] The server organizes the extracted solutions and dynamically adjusts the proposals based on the emotion score using an emotion engine.

[0960] Step 13:

[0961] The server sends the generated proposal to the terminal.

[0962] Step 14:

[0963] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[0964] Step 15:

[0965] The user enters feedback on the provided suggestions into a chat interface.

[0966] Step 16:

[0967] The device sends the user's feedback to the server.

[0968] Step 17:

[0969] The server receives the feedback and stores it in a database.

[0970] Step 18:

[0971] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[0972] Step 19:

[0973] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[0974] Step 20:

[0975] The next time the system is used, the server will use the updated machine learning model and emotional data to provide more accurate, optimized support to the user.

[0976] Example 2

[0977] 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."

[0978] In recent years, mental health issues have been increasing, creating a demand for effective and prompt support. However, it is difficult to provide an anonymous environment where users can seek advice in a safe and secure environment. Furthermore, there are limited systems that provide appropriate advice based on the user's emotions and the content of their consultation. As a result, the accuracy and effectiveness of responses to users' consultations are often insufficient.

[0979] 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.

[0980] In this invention, the server includes a means for a user to input the consultation content in text format, a means for receiving the consultation content in the server and performing an initial analysis using natural language processing, and a means for dynamically adjusting the content of suggestions based on the user's emotion score. This provides an environment where users can consult anonymously with peace of mind, and by using an emotion engine, it becomes possible to provide personalized support optimized for the user's emotions in real time.

[0981] "User" refers to an individual who accesses the system and inputs the details of their consultation.

[0982] "Terminal" refers to a device used by a user, such as a personal computer or smartphone.

[0983] "Server" refers to a central computer system that receives and analyzes consultation content, generates proposals, and provides them to users.

[0984] "Natural language processing" refers to the technology for analyzing input text data and understanding its meaning and emotions.

[0985] "Initial analysis" refers to the process of first analyzing the text of the input consultation content, calculating the emotion score, and classifying topics.

[0986] An "emotion engine" refers to an algorithm or model that calculates an emotional score from text data entered by the user and recognizes their emotional state.

[0987] "Emotion score" refers to a numerical value that indicates the user's emotional state, calculated based on the input text data.

[0988] "Topic classification" refers to the process of sorting the input consultation content into appropriate categories based on its content.

[0989] "Database" refers to a storage device or system for storing past consultation details and feedback data.

[0990] A "machine learning model" refers to an algorithm or program that is trained using collected data to provide more accurate support for the next consultation.

[0991] "Suggestion" refers to advice or solutions generated by the server and provided to the user.

[0992] "Feedback" refers to the opinions and evaluations that users enter regarding the suggestions.

[0993] "Dynamic adjustment" refers to changing the suggestions in real time in response to changes in the user's emotional score, etc.

[0994] "Emergency Support" refers to fast, special support provided when a user's inquiry requires urgent attention.

[0995] "Alert" means a warning or notification generated when it is determined that emergency support is required.

[0996] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system is characterized by incorporating an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing will be explained below.

[0997] User Access and Login

[0998] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[0999] Enter and submit your inquiry

[1000] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[1001] Receiving consultation content and initial analysis

[1002] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The consultation content is also classified into topics and sorted into appropriate categories. Natural language processing technology used includes Python's NLTK and SpaCy.

[1003] Historical database queries

[1004] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions. It then organizes the extracted solutions and generates proposals to be provided to users. To achieve this, a database management system (e.g., MySQL or PostgreSQL) is used.

[1005] Emotional engine regulation

[1006] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score, providing advice and solutions optimized for the user's emotions. The emotion engine uses emotion analysis models such as BERT and RoBERTa.

[1007] Proposal generation and delivery

[1008] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[1009] Receive feedback and learn

[1010] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[1011] Specific examples

[1012] Example 1: Work stress

[1013] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[1014] The terminal transmits the consultation contents to the server.

[1015] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[1016] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[1017] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[1018] The server sends the proposal to the device.

[1019] The device displays the suggestions to the user.

[1020] Users try out the proposed methods and provide feedback.

[1021] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[1022] Example 2: Relationship problems

[1023] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[1024] The terminal transmits the consultation contents to the server.

[1025] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[1026] The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[1027] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[1028] The server sends the proposal to the device.

[1029] The device displays the suggestions to the user.

[1030] The user implements the proposed method and provides feedback.

[1031] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[1032] This invention provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, personalized support optimized for the user's emotions is provided in real time. Continuous feedback constantly improves the accuracy and effectiveness of the system.

[1033] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1034] Step 1:

[1035] A user accesses the system's website or application using a terminal such as a personal computer or smartphone.

[1036] Input: Website URL or application icon

[1037] Output: Login screen displayed

[1038] What happens: The user enters a URL in a web browser or taps an application to launch it. The device renders and displays a login screen to the user.

[1039] Step 2:

[1040] The device will display a login screen and prompt you to enter your username and password.

[1041] Input: None

[1042] Output: Username and password input fields displayed

[1043] What it does: The terminal renders a login form with username and password input fields.

[1044] Step 3:

[1045] The user enters their username and password and clicks the submit button.

[1046] Input: Username and Password

[1047] Output: Send authentication information to the server

[1048] How it works: The user enters information on the keyboard and clicks the login button. The device sends the information to the server.

[1049] Step 4:

[1050] Verify the authentication information received by the server.

[1051] Input: Username and Password

[1052] Output: Authentication result

[1053] What it does: The server queries its database to verify that the username and password submitted match.

[1054] Step 5:

[1055] If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[1056] Input: Authentication result (success)

[1057] Output: Display URL for dashboard or chat interface

[1058] How it works: The server generates a session ID, saves the session data, and responds with a URL for the dashboard or chat interface, which the device receives and displays to the user.

[1059] Step 6:

[1060] The user enters the content of the consultation in text format into the chat interface.

[1061] Input: Text of consultation

[1062] Output: Consultation content data

[1063] Action: The user enters text using a keyboard or touchscreen and clicks the send button. The device sends the entered text data to the server.

[1064] Step 7:

[1065] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[1066] Input: Consultation content data

[1067] Output: Initial analysis results (tokenized text and sentiment scores)

[1068] How it works: The server uses Python libraries such as NLTK and SpaCy to tokenize and morphologically analyze the text and calculate a sentiment score.

[1069] Step 8:

[1070] The server categorizes the consultation content into topics based on the initial analysis results.

[1071] Input: Initial analysis results

[1072] Output: Topic classification results

[1073] How it works: The server uses topic models (LDA or LSI) to classify text data into categories.

[1074] Step 9:

[1075] The server queries a database of past consultations and extracts similar consultations and their solutions.

[1076] Input: Topic classification results

[1077] Output: Similar issues and solutions

[1078] How it works: The server generates an SQL query and sends it to the database to retrieve similar queries and their solutions.

[1079] Step 10:

[1080] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score.

[1081] Input: Similar issues and solutions, sentiment scores

[1082] Output: Dynamically adjusted suggestions

[1083] How it works: The server adjusts the tone and content of the solution it presents based on the sentiment score.

[1084] Step 11:

[1085] The server sends the generated proposal to the terminal.

[1086] Input: Dynamically adjusted suggestions

[1087] Output: Proposal content data

[1088] How it works: The server sends the suggestions as an HTTP response, which the device receives and renders in the chat interface.

[1089] Step 12:

[1090] Users enter feedback on the suggestions provided into a chat interface.

[1091] Input: Feedback text

[1092] Output: Feedback data

[1093] Action: The user enters feedback in text format and clicks the send button. The device sends the feedback data to the server.

[1094] Step 13:

[1095] The server accumulates user feedback in a database and trains a machine learning model.

[1096] Input: Feedback data

[1097] Output: An updated machine learning model

[1098] How it works: The server stores the feedback data in a database and periodically retrains the machine learning model.

[1099] Step 14:

[1100] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[1101] Input: User's historical sentiment score

[1102] Output: Personalized support

[1103] How it works: The server analyzes historical data of sentiment scores and applies an algorithm to track long-term sentiment trends.

[1104] (Application example 2)

[1105] 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."

[1106] While existing mental health support systems provide an environment where users can seek advice anonymously and with peace of mind, they do not adequately respond appropriately to the user's emotions. Furthermore, they lacked mechanisms for effectively utilizing user feedback to improve the accuracy of future consultations, making it difficult to provide optimal support tailored to each user's individual situation. Furthermore, there were no systems that could suggest foods or beverages based on the content of mental health consultations, meaning there was a lack of a means to provide immediate, specific support for specific emotions or physical conditions.

[1107] 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.

[1108] In this invention, the server includes: a means for a user to input a text-based consultation content; a means for receiving the consultation content in the server and performing an initial analysis using natural language processing; a means for topic-classifying the consultation content based on the results of the initial analysis; a means for querying a database of past consultation content to extract similar consultation content and solutions; a means for organizing the extracted solutions and generating a proposal content to be provided to the user; a means for sending the generated proposal content to the user's terminal and displaying it to the user; a means for receiving feedback from the user and storing the feedback in a database; a means for training a machine learning model based on the accumulated data to generate more accurate support for the next consultation; a means for dynamically optimizing food and beverage suggestions based on the user's consultation content and emotion score; a means for generating food and beverage suggestions based on the user's emotion using a generative AI model; and a means for generating an alert if the generated proposal content determines that emergency support is required. This allows for the provision of proposal content optimized to the user's emotion, enabling immediate food and beverage suggestions tailored to specific emotions or physical conditions. Furthermore, by utilizing user feedback, the system's accuracy can be improved, enabling more personalized support to be provided.

[1109] "User" refers to an individual who uses the System.

[1110] "Text-format consultation content" refers to text information that a user inputs into the system, including consultations and questions.

[1111] "Server" refers to a computing system that receives, processes, and transmits data.

[1112] "Natural language processing" is a technology that uses computers to understand, analyze, and generate human language.

[1113] "Initial analysis" is the process of first analyzing the input text data to understand its basic structure and meaning.

[1114] "Topic classification" refers to the process of sorting text data into specific subjects or categories.

[1115] "Past consultation content database" refers to a data bank that stores consultation content and solutions that users have previously entered into the system.

[1116] "Similar consultation contents and their solutions" refers to consultation contents and their solutions that are highly similar to the current consultation contents and past consultation contents in the database.

[1117] "Suggestions" refers to the advice or solutions the system provides to the user.

[1118] "User Device" refers to the electronic device (e.g., smartphone, tablet, or PC) used by a User to access the System.

[1119] "Feedback" refers to the evaluations and opinions that users give regarding the suggestions provided.

[1120] A "machine learning model" refers to an algorithm that automatically learns from accumulated data and makes predictions and classifications.

[1121] "More precise support" refers to assistance that responds appropriately and effectively to the user's individual situation.

[1122] An "emotion score" is an index that numerically evaluates a user's emotional state based on their text input.

[1123] "Dynamically optimized food and drink" refers to food and drink suggestions that are tailored based on the user's emotions and physical condition.

[1124] "Generative AI model" refers to an artificial intelligence model used to perform a generative task.

[1125] "Emergency Support" means an emergency response required immediately by a User.

[1126] An "alert" is a notification that notifies you of an emergency or important information.

[1127] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. The system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions.

[1128] System configuration

[1129] This system includes a user device, a server, a database, a natural language processing engine, an emotion engine, and a machine learning model. The user device refers to an electronic device such as a smartphone, tablet, or PC, and the server is a computer system for receiving, processing, and transmitting data.

[1130] Program processing

[1131] Login and Access

[1132] A user accesses the system's website or application using their own device and enters their username and password to log in. This authentication information is sent to the server, and once the server has successfully authenticated, the user is provided with a chat interface for consultation.

[1133] Enter and submit your inquiry

[1134] The user enters the content of the consultation in text format into the chat interface, and this content is sent to the server.

[1135] Initial analysis and sentiment analysis

[1136] The server uses natural language processing technology to perform an initial analysis of the received consultation content, and then calculates an emotion score using an emotion engine, which allows the user's emotional state to be numerically evaluated.

[1137] Topic classification and historical data reference

[1138] Based on the analysis results, the consultation content is classified into topics, and similar consultation content and solutions are extracted by referring to a database of past consultation content.

[1139] Proposal generation and optimization

[1140] The server generates recommendations based on the extracted solutions and dynamically optimizes them based on the emotion score, suggesting foods and drinks that correspond to the user's emotions and physical condition.

[1141] View suggestions and get feedback

[1142] The generated suggestions are sent back to the user's device and displayed to the user. The user checks the suggestions and enters feedback. This feedback is sent to the server and stored in a database.

[1143] Training a machine learning model

[1144] The server uses the accumulated feedback to train a machine learning model and generate more accurate support for the next consultation.

[1145] Hardware and software used

[1146] Hardware: User devices (smartphones, tablets, PCs), servers (high-performance computing systems)

[1147] Software: Natural language processing engines (e.g., SpaCy, NLTK), emotion engines (e.g., Hugging Face's transformers library), machine learning models (e.g., Scikit-learn, TensorFlow)

[1148] Specific examples

[1149] Example 1:

[1150] If a user types, "I've been stressed out lately and have lost my appetite.":

[1151] The system suggests, "Would you like some relaxing herbal tea?"

[1152] The user orders the suggested herbal tea and enters feedback as "That was a good suggestion, I feel better."

[1153] Prompt Sentence Examples

[1154] "The user has recently been feeling stressed and has lost their appetite. Please suggest suitable foods and drinks. Use your historical database and emotion engine to generate the best suggestions."

[1155] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1156] Specific processing flow of the program

[1157] Step 1: User Login and Access

[1158] Input: The user enters their username and password from their device (smartphone, tablet, computer).

[1159] Data processing: The device sends the entered authentication information (username and password) to the server.

[1160] Data calculation: The server checks the authentication information against the database.

[1161] Output: If authentication is successful, the server starts a session and presents the user with a chat interface for consultation.

[1162] Step 2: Enter and submit your inquiry

[1163] Input: The user enters the content of their inquiry in text format into the chat interface.

[1164] Data processing: The device sends the consultation details entered to the server.

[1165] Data calculation: None (simple data transmission).

[1166] Output: The consultation content is received by the server.

[1167] Step 3: Initial analysis and sentiment analysis

[1168] Input: Text data of the consultation received by the server.

[1169] Data processing: The server performs initial analysis of the text data using a natural language processing engine (e.g., SpaCy, NLTK).

[1170] Data calculation: An emotion engine (e.g., Hugging Face transformers library) is used to calculate an emotion score and numerically evaluate the emotional state of the consultation.

[1171] Output: Analysis results of the text data and sentiment scores.

[1172] Step 4: Topic classification and historical data reference

[1173] Input: Initial analysis results and sentiment scores.

[1174] Data processing: The server categorizes the consultation content into topics based on the analysis results.

[1175] Data calculation: Query the database of past consultations to extract similar consultations and their solutions.

[1176] Output: Topic classification results, and similar consultations and their solutions.

[1177] Step 5: Generate and optimize proposals

[1178] Input: Similar consultations and their solutions, sentiment scores.

[1179] Data processing: The server generates suggestions using the generative AI model.

[1180] Data computation: Dynamically optimizes recommendations based on emotion scores, suggesting foods and drinks that correspond to the user's emotions and physical condition.

[1181] Output: Dynamically optimized suggestions.

[1182] Step 6: View your proposal and get feedback

[1183] Input: Dynamically optimized suggestions.

[1184] Data processing: The server sends the proposal to the user's device.

[1185] Data calculation: None (simple data transmission).

[1186] Output: The suggestion is displayed on the user's device.

[1187] Step 7: Capture and store feedback

[1188] Input: User feedback.

[1189] Data processing: The device sends feedback to the server.

[1190] Data calculation: The server stores the received feedback in a database.

[1191] Output: Feedback stored in a database.

[1192] Step 8: Train the machine learning model

[1193] Input: Accumulated feedback data.

[1194] Data processing: The server inputs the feedback data into the machine learning model.

[1195] Data Computing: Using machine learning models (e.g., Scikit-learn, TensorFlow) to train the next consultation to improve its accuracy.

[1196] Output: An improved machine learning model.

[1197] Specific examples

[1198] Example 1: If a user enters, "I've been feeling stressed lately and have no appetite," the system will suggest, "How about some herbal tea to help you relax?" and the user will confirm and enter their feedback.

[1199] Example prompt sentence:

[1200] "The user has recently been feeling stressed and has lost their appetite. Please suggest suitable foods and drinks. Use your historical database and emotion engine to generate the best suggestions."

[1201] 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.

[1202] 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.

[1203] 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.

[1204] [Third embodiment]

[1205] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1206] 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.

[1207] 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).

[1208] 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.

[1209] 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.

[1210] 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).

[1211] 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.

[1212] 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.

[1213] 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.

[1214] 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.

[1215] 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.

[1216] 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."

[1217] This invention provides a system that allows users to receive mental health consultations and appropriate support without having to deal with face-to-face consultations. The system is composed of multiple means. Each means and its specific processing will be described below.

[1218] User Access and Login

[1219] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[1220] Enter and submit your inquiry

[1221] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[1222] Receiving consultation content and initial analysis

[1223] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, it performs sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[1224] Historical database queries

[1225] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[1226] Proposal generation and delivery

[1227] The server sends the generated suggestions to the device, which displays them in the chat interface and allows the user to review the advice or solution provided.

[1228] Receive feedback and learn

[1229] The user inputs feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation.

[1230] Specific examples

[1231] Example 1: Work stress

[1232] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[1233] The terminal transmits the consultation contents to the server.

[1234] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigning a low emotional score and classifying the topic as "work stress."

[1235] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[1236] The server generates the proposal and sends it to the device.

[1237] The device displays the suggestions to the user.

[1238] Users try out the proposed methods and provide feedback.

[1239] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[1240] Example 2: Relationship problems

[1241] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[1242] The terminal transmits the consultation contents to the server.

[1243] The server analyzes text data containing the keywords "colleague" and "poor relationship" and classifies sentiment scores and topics.

[1244] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[1245] The server generates the proposal and sends it to the device.

[1246] The device displays the suggestions to the user.

[1247] The user reviews the solution and provides feedback.

[1248] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[1249] This system provides users with an environment where they can seek advice anonymously and with peace of mind, and continuous feedback improves the accuracy of the system.

[1250] The processing flow will be explained below.

[1251] Step 1:

[1252] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[1253] Step 2:

[1254] The device displays a login screen and the user enters their username and password.

[1255] Step 3:

[1256] The terminal transmits the entered authentication information to the server.

[1257] Step 4:

[1258] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[1259] Step 5:

[1260] The server displays a dashboard or chat interface to the authenticated user.

[1261] Step 6:

[1262] The user enters the content of the consultation in text format into the chat interface.

[1263] Step 7:

[1264] The terminal transmits the input consultation content to the server.

[1265] Step 8:

[1266] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[1267] Step 9:

[1268] As part of the initial analysis, the server performs sentiment analysis to calculate a sentiment score for the consultation content, and also performs topic classification to assign the content to the appropriate category.

[1269] Step 10:

[1270] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[1271] Step 11:

[1272] The server organizes the extracted solutions and generates proposals.

[1273] Step 12:

[1274] The server sends the generated proposal to the terminal.

[1275] Step 13:

[1276] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[1277] Step 14:

[1278] The user enters feedback on the provided suggestions into a chat interface.

[1279] Step 15:

[1280] The device sends the user's feedback to the server.

[1281] Step 16:

[1282] The server receives the feedback and stores it in a database.

[1283] Step 17:

[1284] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[1285] Step 18:

[1286] The next time the user uses the system, the server will use the updated machine learning model to provide more personalized support to the user.

[1287] Example 1

[1288] 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."

[1289] Conventional mental health consultation systems have difficulty providing appropriate support to users who wish to avoid in-person consultations. Furthermore, these systems lack sufficient mechanisms for effectively utilizing user feedback to improve the accuracy of subsequent consultations. Furthermore, they often lack the means to properly analyze the content of the consultation and provide appropriate solutions. To address these issues, the present invention provides a system that allows users to safely seek mental health consultations and receive appropriate support while avoiding in-person consultations.

[1290] 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.

[1291] In this invention, the server includes: means for a user to input a consultation content in text format; means for a terminal to transmit the consultation content to the server; means for the server to receive the consultation content and perform an initial analysis using natural language processing technology; means for classifying the consultation content by topic based on the results of the initial analysis; means for querying a database of past consultation content and extracting similar consultation content and solutions; means for organizing the extracted solutions and generating a proposal content to be provided to the user; means for transmitting the generated proposal content to the user's terminal and displaying it to the user; means for receiving feedback from the user and storing the feedback in a database; and means for training a machine learning model based on the stored data to generate more accurate support for the next consultation. This allows users to receive mental health consultation anonymously and with peace of mind, and by utilizing continuous feedback, the system can provide more accurate support for subsequent consultations.

[1292] "User" refers to a person who uses the system to seek mental health advice.

[1293] "Terminal" refers to a device, such as a personal computer or smartphone, that a user uses to access the system and enter consultation details.

[1294] "Server" refers to a computer system that processes the inquiry content sent by the user and generates and provides proposal content.

[1295] "Consultation content" refers to text information entered by users about questions and concerns about mental health.

[1296] "Natural language processing technology" refers to technology that analyzes text data, calculates sentiment scores, and performs topic classification.

[1297] "Initial analysis" refers to the initial analysis process that the server performs after receiving the consultation content using natural language processing technology.

[1298] "Topic classification" refers to the process of categorizing consultation content into specific categories or themes based on the results of the initial analysis.

[1299] A "database" refers to a collection of information that stores and manages past consultation details and their solutions.

[1300] A "machine learning model" refers to an algorithm or system that is trained to provide more accurate support for the next consultation based on feedback accumulated in a database.

[1301] "Sentiment analysis" refers to the technology of calculating emotional trends and emotional scores from text data of consultation content.

[1302] "Emotion score" refers to the numerical representation of the intensity and tendency of emotions contained in the consultation content as a result of emotion analysis.

[1303] "Proposal" refers to advice or solutions generated by the server and provided to the user.

[1304] "Feedback" refers to the ratings and comments that users enter in response to suggestions.

[1305] This invention relates to a system that allows users to receive mental health consultations and appropriate support in a safe environment while avoiding face-to-face consultations. The specific configuration and operation of this system will be described below.

[1306] User Access and Login

[1307] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The entered authentication information is sent from the device to the server. The server verifies the authentication information, and if authentication is successful, starts a session and displays a dashboard and chat interface to the user.

[1308] Enter and submit your inquiry

[1309] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[1310] Receiving consultation content and initial analysis

[1311] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology (e.g., Google Cloud Natural Language API, AWS Comprehend, etc.). The analysis includes sentiment analysis and topic classification. Sentiment analysis calculates an emotional score for the consultation content, and topic classification classifies the consultation content into specific categories.

[1312] Historical database queries

[1313] Based on the initial analysis results, the server queries a database of past consultations (e.g., MySQL, PostgreSQL, etc.) to extract similar consultations and their solutions. The server then organizes these extracted solutions and generates proposals to be provided to users.

[1314] Proposal generation and delivery

[1315] The server generates suggestions and sends them to the device, which displays them in a chat interface, allowing the user to review the advice and solutions provided.

[1316] Receive feedback and learn

[1317] The user enters feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores the feedback in a database and trains a machine learning model to provide more accurate support for the next consultation.

[1318] Specific examples

[1319] Example 1: Work stress

[1320] 1. A user says, "I haven't been able to sleep lately because of work stress."

[1321] 2. The device sends the consultation details to the server.

[1322] 3. The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigns a low emotional score, and classifies the topic as "work stress."

[1323] 4. The server queries the historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[1324] 5. The server generates the proposal and sends it to the device.

[1325] 6. The device displays the suggestions to the user.

[1326] 7. The user tries out the proposed solution and provides feedback.

[1327] 8. The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[1328] Example 2: Relationship problems

[1329] 1. A user comes to us with a problem: "I'm having trouble with my coworkers."

[1330] 2. The device sends the consultation details to the server.

[1331] 3. The server analyzes text data containing the keywords "colleague" and "relationship not going well" and classifies sentiment scores and topics.

[1332] 4. The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[1333] 5. The server generates the proposal and sends it to the device.

[1334] 6. The device displays the suggestions to the user.

[1335] 7. The user reviews the solution and provides feedback.

[1336] 8. The device sends the feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[1337] Prompt Sentence Examples

[1338] "I haven't been able to sleep lately because of work stress. What should I do?"

[1339] "I'm having trouble with my coworkers. How can I improve it?"

[1340] In this way, users can seek mental health advice anonymously and with peace of mind, and by utilizing continuous feedback, the system will be able to provide more accurate support for subsequent consultations.

[1341] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1342] Step 1:

[1343] A user accesses the system's website or application using a device such as a personal computer or smartphone. The input is a username and password, and the device displays a login screen, which collects the authentication information. The output is the authentication information entered by the user.

[1344] Step 2:

[1345] The device sends the entered authentication information to the server. The input is the username and password entered by the user, and the output is the authentication information sent to the server. The device securely transmits the authentication information using the SSL / TLS protocol.

[1346] Step 3:

[1347] The server queries the user database based on the received authentication information to verify authentication. The input is the received authentication information, and the output is whether the authentication was successful. If authentication is successful, the server generates a session ID and generates information to display the dashboard and chat interface.

[1348] Step 4:

[1349] The user inputs the content of the consultation in text format into the chat interface. The input is the text of the consultation content, and the output is the input text data.

[1350] Step 5:

[1351] The terminal transmits the input consultation content to the server. The input is text data entered by the user, and the output is text data transmitted to the server. The terminal transmits the data using a communication protocol.

[1352] Step 6:

[1353] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. The input is the received text data, and the output is the analysis results. The server performs sentiment analysis, calculates sentiment scores, and performs topic classification.

[1354] Step 7:

[1355] The server queries a database of past consultations based on the initial analysis results. The input is the initial analysis results, and the output is similar consultations and their solutions. The server queries the database and organizes the extracted solutions.

[1356] Step 8:

[1357] The server generates a proposal to be provided to the user based on the extracted solution. The input is the extracted solution, and the output is the proposal. The server sends the generated proposal to the terminal.

[1358] Step 9:

[1359] The terminal displays the suggestions in a chat interface, and the user confirms the offered advice or solution. The input is the suggestion sent from the server, and the output is the suggestion displayed in the chat interface.

[1360] Step 10:

[1361] The user inputs feedback on the provided suggestion. The input is the feedback, and the output is the input feedback.

[1362] Step 11:

[1363] The terminal sends the user's feedback to the server. The input is the feedback content, and the output is the feedback sent to the server. The terminal transmits the data using a communication protocol.

[1364] Step 12:

[1365] The server accumulates the feedback in a database and trains a machine learning model. The input is the feedback content, and the output is the trained machine learning model. This makes it possible to provide more accurate support for the next consultation.

[1366] (Application example 1)

[1367] 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."

[1368] Mental health care for drivers in autonomous vehicles is an important issue for ensuring safe driving and reducing driver stress. However, conventional systems have no way for drivers to receive appropriate support in real time while avoiding face-to-face consultations. This has led to a lack of mental health follow-up in emergencies while driving.

[1369] 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.

[1370] In this invention, the server includes means for a user to input a consultation content in text format, means for receiving the consultation content in the server and performing an initial analysis using natural language processing, means for topic categorizing the consultation content based on the results of the initial analysis, means for consulting a database of past consultation content and extracting similar consultation contents and their solutions, means for organizing the extracted solutions and generating a proposal content to be provided to the user, means for sending the generated proposal content to the user's terminal and displaying it to the user, means for receiving feedback from the user and storing the feedback in a database, means for training a machine learning model based on the stored data and generating more accurate support for the next consultation, means for providing an interface for providing mental health care in an autonomous vehicle, and means for monitoring the driver's mental state while driving and providing support in real time, thereby enabling the driver to receive mental health care in real time in an autonomous vehicle.

[1371] "User" refers to a driver who uses the system to seek mental health advice.

[1372] "Text consultation content" refers to the string of characters that users enter into the system, and the content represents mental health issues and concerns.

[1373] The "server" is the computer at the core of the system, which receives inquiries from users and analyzes them and provides support suggestions.

[1374] "Natural language processing" is a technology that allows a computer to analyze text data entered by a user and understand its meaning and emotions.

[1375] "Initial analysis" is the process of first analyzing the consultation content using natural language processing technology and extracting emotion scores and keywords.

[1376] "Topic classification" refers to classifying the content of a consultation into multiple categories based on the results of the initial analysis.

[1377] The "past consultation content database" is a database that stores consultation content and solutions that have been accumulated in the system up to now.

[1378] "Similar consultation content" refers to cases in the database of past consultation content that are similar in content and emotion to the current consultation content.

[1379] "Solutions" are advice or measures provided to resolve a user's mental health issues based on past consultations.

[1380] The "proposal content" is a support proposal generated for the current consultation content based on the extracted solution.

[1381] "User's device" means a device used by a user, including a smartphone, tablet, or in-vehicle interface.

[1382] "Feedback" refers to the opinions and evaluations that users give in response to suggestions, and is used to improve the accuracy of the system.

[1383] A "machine learning model" is an algorithm that learns from large amounts of data and generates more accurate support for the next consultation.

[1384] An "autonomous vehicle" is a vehicle that automatically performs driving operations and provides a place for the driver to receive mental health care.

[1385] An "interface" is an operating means, such as a screen or voice input system, that allows a user to interact with a system.

[1386] "Mental state monitoring" is the process of monitoring a driver's emotions and stress levels and collecting data in real time.

[1387] This invention is a system that allows users to receive mental health consultations and appropriate support while in an autonomous vehicle. The specific configuration and procedure are shown below.

[1388] 1. System Configuration

[1389] Main components:

[1390] User devices: devices such as touchscreens in vehicles and smartphones

[1391] Server: Cloud-based computing cluster

[1392] Natural language processing models, such as BERT for Hugging Face

[1393] Database: A database for storing past consultations and feedback

[1394] 2. User Access and Login

[1395] Users access the system using a touchscreen in their vehicle or a smartphone, enter their username and password, and the device sends the credentials to a server. After authentication, the session begins and the user is presented with a dashboard and chat interface.

[1396] 3. Enter and submit your inquiry

[1397] The user inputs the text of the consultation into the chat interface, and the device sends the inputted consultation content to the server.

[1398] 4. Receiving consultation content and initial analysis

[1399] When the server receives the consultation content, it performs an initial analysis of the text data using a natural language processing model. As part of the analysis, it performs a sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[1400] 5. Query the database and generate suggestions

[1401] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions, and then organizes the extracted solutions to generate proposals to be provided to the user.

[1402] 6. Providing suggestions and feedback

[1403] The server generates suggestions and sends them to the user's device, which displays them in the chat interface. The user reviews the advice and solutions provided and enters feedback on the results. The device then sends the feedback to the server, which stores it in a database.

[1404] 7. Feedback learning and model training

[1405] The server uses the accumulated feedback to train a machine learning model and provide more accurate support for the next consultation.

[1406] 8. Operation in autonomous vehicles

[1407] The system can monitor the driver's mental state while driving in an autonomous vehicle and provide appropriate support in real time. For example, if the driver feels anxious on the highway, the system can suggest deep breathing or meditation to relieve tension.

[1408] Specific examples

[1409] 1. Example 1: Highway anxiety

[1410] Consultation: "Recently, I've been feeling anxious when driving on the highway."

[1411] Suggestion: "To reduce highway anxiety, try listening to appropriate music and taking deep breaths."

[1412] Feedback: "I tried the deep breathing exercises you suggested and my anxiety has decreased a little."

[1413] 2. Example 2: Commuting stress

[1414] Consultation: "I always feel stressed during my morning commute."

[1415] Suggestion: "Trying a short meditation session before you start driving may help reduce stress."

[1416] Feedback: "I tried meditation and it made me feel a little less stressed in the mornings."

[1417] In this way, the driver can receive real-time mental health care in the vehicle, providing a safe and comfortable driving environment.

[1418] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1419] Step 1:

[1420] The user enters their username and password using the vehicle's touchscreen or smartphone.

[1421] Input: Username, Password

[1422] Output: Credentials

[1423] Specific operation:

[1424] The user operates the login screen on the vehicle's touchscreen and enters their username and password.

[1425] Step 2:

[1426] The terminal transmits the entered authentication information to the server, and the server performs authentication.

[1427] Input: Credentials

[1428] Output: Authentication result (success / failure)

[1429] Specific operation:

[1430] The device sends the user's input information to the server via encrypted communication such as SSL, and the server compares it with a database and returns the authentication result.

[1431] Step 3:

[1432] If authentication is successful, the server starts a session and presents the user with a dashboard and chat interface.

[1433] Input: Authentication result (success)

[1434] Output: Dashboard, chat interface

[1435] Specific operation:

[1436] The server generates a session ID and sends it to the user's device, which receives it and displays the dashboard and chat interface.

[1437] Step 4:

[1438] The user inputs the content of the consultation in text format into the chat interface, and the terminal transmits the content to the server.

[1439] Input: Consultation content (text format)

[1440] Output: Sending the consultation details to the server

[1441] Specific operation:

[1442] A user types "Lately, I feel anxious when I drive on the highway" into the chat interface and clicks the send button. The device then sends this text to the server.

[1443] Step 5:

[1444] The server receives the consultation content and performs an initial analysis using a natural language processing model.

[1445] Input: Consultation content (text format)

[1446] Output: Initial analysis results (sentiment score, topic classification)

[1447] Specific operation:

[1448] The server performs sentiment analysis using Hugging Face's BERT model and other tools to analyze the content of the consultation. The analysis results include "anxiety (NEGATIVE)."

[1449] Step 6:

[1450] The server queries a database of past consultation contents based on the results of the initial analysis and extracts similar consultation contents and their solutions.

[1451] Input: Initial analysis results (sentiment score, topic classification)

[1452] Output: Similar inquiries and solutions

[1453] Specific operation:

[1454] The server uses the analysis results to query a database and extract similar consultations and past solutions (e.g., deep breathing, meditation, etc.).

[1455] Step 7:

[1456] The server organizes the extracted solutions and generates proposals to be presented to the user.

[1457] Input: Similar inquiries and solutions

[1458] Output: Proposal

[1459] Specific operation:

[1460] The server organizes solutions and generates suggestions such as, "To reduce anxiety on the highway, try listening to appropriate music or taking deep breaths."

[1461] Step 8:

[1462] The server sends the generated proposals to the user's device, which then displays them to the user.

[1463] Input: Proposal

[1464] Output: what is displayed to the user

[1465] Specific operation:

[1466] The server transmits the proposal to the terminal, which displays the proposal on the chat interface.

[1467] Step 9:

[1468] The user tries the proposed method and inputs feedback on the results, which is then sent to the server.

[1469] Input: Feedback (text format)

[1470] Output: Send feedback

[1471] Specific operation:

[1472] The user types, "I tried the suggested deep breathing exercises and my anxiety has decreased a little," and submits the feedback. The device then sends this feedback to the server.

[1473] Step 10:

[1474] The server receives the feedback, stores it in a database, and trains a machine learning model to provide more accurate support for the next consultation.

[1475] Input: Feedback

[1476] Output: A trained machine learning model

[1477] Specific operation:

[1478] The server stores the feedback in a database and uses this data to update and train the machine learning model, improving the accuracy of support for the next consultation.

[1479] 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.

[1480] The present invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing are explained below.

[1481] User Access and Login

[1482] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[1483] Enter and submit your inquiry

[1484] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[1485] Receiving consultation content and initial analysis

[1486] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The server also categorizes the consultation content into topics and assigns them to appropriate categories.

[1487] Historical database queries

[1488] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[1489] Emotional engine regulation

[1490] The server uses an emotion engine to dynamically adjust its suggestions based on the user's emotion score, providing advice and solutions that are optimized for the user's emotions.

[1491] Proposal generation and delivery

[1492] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[1493] Receive feedback and learn

[1494] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[1495] Specific examples

[1496] Example 1: Work stress

[1497] A user complains, "I haven't been able to sleep lately because of work stress."

[1498] The terminal transmits the consultation contents to the server.

[1499] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[1500] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[1501] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[1502] The server sends the proposal to the device.

[1503] The device displays the suggestions to the user.

[1504] Users try out the proposed methods and provide feedback.

[1505] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[1506] Example 2: Relationship problems

[1507] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[1508] The terminal transmits the consultation contents to the server.

[1509] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[1510] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[1511] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[1512] The server sends the proposal to the device.

[1513] The device displays the suggestions to the user.

[1514] The user implements the proposed method and provides feedback.

[1515] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[1516] This system provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, it provides personalized support optimized to the user's emotions.Continuous feedback constantly improves the system's accuracy, resulting in more effective support.

[1517] The processing flow will be explained below.

[1518] Step 1:

[1519] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[1520] Step 2:

[1521] The device displays a login screen and the user enters their username and password.

[1522] Step 3:

[1523] The terminal transmits the entered authentication information to the server.

[1524] Step 4:

[1525] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[1526] Step 5:

[1527] The server displays a dashboard or chat interface to the authenticated user.

[1528] Step 6:

[1529] The user enters the content of the consultation in text format into the chat interface.

[1530] Step 7:

[1531] The terminal transmits the input consultation content to the server.

[1532] Step 8:

[1533] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[1534] Step 9:

[1535] As part of the initial analysis, the server uses an emotion engine to calculate an emotion score and recognize the emotion of the consultation content.

[1536] Step 10:

[1537] The server classifies the consultation topic and assigns it to the appropriate category.

[1538] Step 11:

[1539] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[1540] Step 12:

[1541] The server organizes the extracted solutions and dynamically adjusts the proposals based on the emotion score using an emotion engine.

[1542] Step 13:

[1543] The server sends the generated proposal to the terminal.

[1544] Step 14:

[1545] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[1546] Step 15:

[1547] The user enters feedback on the provided suggestions into a chat interface.

[1548] Step 16:

[1549] The device sends the user's feedback to the server.

[1550] Step 17:

[1551] The server receives the feedback and stores it in a database.

[1552] Step 18:

[1553] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[1554] Step 19:

[1555] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[1556] Step 20:

[1557] The next time the system is used, the server will use the updated machine learning model and emotional data to provide more accurate, optimized support to the user.

[1558] Example 2

[1559] 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."

[1560] In recent years, mental health issues have been increasing, creating a demand for effective and prompt support. However, it is difficult to provide an anonymous environment where users can seek advice in a safe and secure environment. Furthermore, there are limited systems that provide appropriate advice based on the user's emotions and the content of their consultation. As a result, the accuracy and effectiveness of responses to users' consultations are often insufficient.

[1561] 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.

[1562] In this invention, the server includes a means for a user to input the consultation content in text format, a means for receiving the consultation content in the server and performing an initial analysis using natural language processing, and a means for dynamically adjusting the content of suggestions based on the user's emotion score. This provides an environment where users can consult anonymously with peace of mind, and by using an emotion engine, it becomes possible to provide personalized support optimized for the user's emotions in real time.

[1563] "User" refers to an individual who accesses the system and inputs the details of their consultation.

[1564] "Terminal" refers to a device used by a user, such as a personal computer or smartphone.

[1565] "Server" refers to a central computer system that receives and analyzes consultation content, generates proposals, and provides them to users.

[1566] "Natural language processing" refers to the technology for analyzing input text data and understanding its meaning and emotions.

[1567] "Initial analysis" refers to the process of first analyzing the text of the input consultation content, calculating the emotion score, and classifying topics.

[1568] An "emotion engine" refers to an algorithm or model that calculates an emotional score from text data entered by the user and recognizes their emotional state.

[1569] "Emotion score" refers to a numerical value that indicates the user's emotional state, calculated based on the input text data.

[1570] "Topic classification" refers to the process of sorting the input consultation content into appropriate categories based on its content.

[1571] "Database" refers to a storage device or system for storing past consultation details and feedback data.

[1572] A "machine learning model" refers to an algorithm or program that is trained using collected data to provide more accurate support for the next consultation.

[1573] "Suggestion" refers to advice or solutions generated by the server and provided to the user.

[1574] "Feedback" refers to the opinions and evaluations that users enter regarding the suggestions.

[1575] "Dynamic adjustment" refers to changing the suggestions in real time in response to changes in the user's emotional score, etc.

[1576] "Emergency Support" refers to fast, special support provided when a user's inquiry requires urgent attention.

[1577] "Alert" means a warning or notification generated when it is determined that emergency support is required.

[1578] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system is characterized by incorporating an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing will be explained below.

[1579] User Access and Login

[1580] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[1581] Enter and submit your inquiry

[1582] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[1583] Receiving consultation content and initial analysis

[1584] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The consultation content is also classified into topics and sorted into appropriate categories. Natural language processing technology used includes Python's NLTK and SpaCy.

[1585] Historical database queries

[1586] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions. It then organizes the extracted solutions and generates proposals to be provided to users. To achieve this, a database management system (e.g., MySQL or PostgreSQL) is used.

[1587] Emotional engine regulation

[1588] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score, providing advice and solutions optimized for the user's emotions. The emotion engine uses emotion analysis models such as BERT and RoBERTa.

[1589] Proposal generation and delivery

[1590] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[1591] Receive feedback and learn

[1592] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[1593] Specific examples

[1594] Example 1: Work stress

[1595] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[1596] The terminal transmits the consultation contents to the server.

[1597] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[1598] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[1599] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[1600] The server sends the proposal to the device.

[1601] The device displays the suggestions to the user.

[1602] Users try out the proposed methods and provide feedback.

[1603] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[1604] Example 2: Relationship problems

[1605] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[1606] The terminal transmits the consultation contents to the server.

[1607] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[1608] The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[1609] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[1610] The server sends the proposal to the device.

[1611] The device displays the suggestions to the user.

[1612] The user implements the proposed method and provides feedback.

[1613] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[1614] This invention provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, personalized support optimized for the user's emotions is provided in real time. Continuous feedback constantly improves the accuracy and effectiveness of the system.

[1615] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1616] Step 1:

[1617] A user accesses the system's website or application using a terminal such as a personal computer or smartphone.

[1618] Input: Website URL or application icon

[1619] Output: Login screen displayed

[1620] What happens: The user enters a URL in a web browser or taps an application to launch it. The device renders and displays a login screen to the user.

[1621] Step 2:

[1622] The device will display a login screen and prompt you to enter your username and password.

[1623] Input: None

[1624] Output: Username and password input fields displayed

[1625] What it does: The terminal renders a login form with username and password input fields.

[1626] Step 3:

[1627] The user enters their username and password and clicks the submit button.

[1628] Input: Username and Password

[1629] Output: Send authentication information to the server

[1630] How it works: The user enters information on the keyboard and clicks the login button. The device sends the information to the server.

[1631] Step 4:

[1632] Verify the authentication information received by the server.

[1633] Input: Username and Password

[1634] Output: Authentication result

[1635] What it does: The server queries its database to verify that the username and password submitted match.

[1636] Step 5:

[1637] If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[1638] Input: Authentication result (success)

[1639] Output: Display URL for dashboard or chat interface

[1640] How it works: The server generates a session ID, saves the session data, and responds with a URL for the dashboard or chat interface, which the device receives and displays to the user.

[1641] Step 6:

[1642] The user enters the content of the consultation in text format into the chat interface.

[1643] Input: Text of consultation

[1644] Output: Consultation content data

[1645] Action: The user enters text using a keyboard or touchscreen and clicks the send button. The device sends the entered text data to the server.

[1646] Step 7:

[1647] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[1648] Input: Consultation content data

[1649] Output: Initial analysis results (tokenized text and sentiment scores)

[1650] How it works: The server uses Python libraries such as NLTK and SpaCy to tokenize and morphologically analyze the text and calculate a sentiment score.

[1651] Step 8:

[1652] The server categorizes the consultation content into topics based on the initial analysis results.

[1653] Input: Initial analysis results

[1654] Output: Topic classification results

[1655] How it works: The server uses topic models (LDA or LSI) to classify text data into categories.

[1656] Step 9:

[1657] The server queries a database of past consultations and extracts similar consultations and their solutions.

[1658] Input: Topic classification results

[1659] Output: Similar issues and solutions

[1660] How it works: The server generates an SQL query and sends it to the database to retrieve similar queries and their solutions.

[1661] Step 10:

[1662] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score.

[1663] Input: Similar issues and solutions, sentiment scores

[1664] Output: Dynamically adjusted suggestions

[1665] How it works: The server adjusts the tone and content of the solution it presents based on the sentiment score.

[1666] Step 11:

[1667] The server sends the generated proposal to the terminal.

[1668] Input: Dynamically adjusted suggestions

[1669] Output: Proposal content data

[1670] How it works: The server sends the suggestions as an HTTP response, which the device receives and renders in the chat interface.

[1671] Step 12:

[1672] Users enter feedback on the suggestions provided into a chat interface.

[1673] Input: Feedback text

[1674] Output: Feedback data

[1675] Action: The user enters feedback in text format and clicks the send button. The device sends the feedback data to the server.

[1676] Step 13:

[1677] The server accumulates user feedback in a database and trains a machine learning model.

[1678] Input: Feedback data

[1679] Output: An updated machine learning model

[1680] How it works: The server stores the feedback data in a database and periodically retrains the machine learning model.

[1681] Step 14:

[1682] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[1683] Input: User's historical sentiment score

[1684] Output: Personalized support

[1685] How it works: The server analyzes historical data of sentiment scores and applies an algorithm to track long-term sentiment trends.

[1686] (Application example 2)

[1687] 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."

[1688] While existing mental health support systems provide an environment where users can seek advice anonymously and with peace of mind, they do not adequately respond appropriately to the user's emotions. Furthermore, they lacked mechanisms for effectively utilizing user feedback to improve the accuracy of future consultations, making it difficult to provide optimal support tailored to each user's individual situation. Furthermore, there were no systems that could suggest foods or beverages based on the content of mental health consultations, meaning there was a lack of a means to provide immediate, specific support for specific emotions or physical conditions.

[1689] 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.

[1690] In this invention, the server includes: a means for a user to input a text-based consultation content; a means for receiving the consultation content in the server and performing an initial analysis using natural language processing; a means for topic-classifying the consultation content based on the results of the initial analysis; a means for querying a database of past consultation content to extract similar consultation content and solutions; a means for organizing the extracted solutions and generating a proposal content to be provided to the user; a means for sending the generated proposal content to the user's terminal and displaying it to the user; a means for receiving feedback from the user and storing the feedback in a database; a means for training a machine learning model based on the accumulated data to generate more accurate support for the next consultation; a means for dynamically optimizing food and beverage suggestions based on the user's consultation content and emotion score; a means for generating food and beverage suggestions based on the user's emotion using a generative AI model; and a means for generating an alert if the generated proposal content determines that emergency support is required. This allows for the provision of proposal content optimized to the user's emotion, enabling immediate food and beverage suggestions tailored to specific emotions or physical conditions. Furthermore, by utilizing user feedback, the system's accuracy can be improved, enabling more personalized support to be provided.

[1691] "User" refers to an individual who uses the System.

[1692] "Text-format consultation content" refers to text information that a user inputs into the system, including consultations and questions.

[1693] "Server" refers to a computing system that receives, processes, and transmits data.

[1694] "Natural language processing" is a technology that uses computers to understand, analyze, and generate human language.

[1695] "Initial analysis" is the process of first analyzing the input text data to understand its basic structure and meaning.

[1696] "Topic classification" refers to the process of sorting text data into specific subjects or categories.

[1697] "Past consultation content database" refers to a data bank that stores consultation content and solutions that users have previously entered into the system.

[1698] "Similar consultation contents and their solutions" refers to consultation contents and their solutions that are highly similar to the current consultation contents and past consultation contents in the database.

[1699] "Suggestions" refers to the advice or solutions the system provides to the user.

[1700] "User Device" refers to the electronic device (e.g., smartphone, tablet, or PC) used by a User to access the System.

[1701] "Feedback" refers to the evaluations and opinions that users give regarding the suggestions provided.

[1702] A "machine learning model" refers to an algorithm that automatically learns from accumulated data and makes predictions and classifications.

[1703] "More precise support" refers to assistance that responds appropriately and effectively to the user's individual situation.

[1704] An "emotion score" is an index that numerically evaluates a user's emotional state based on their text input.

[1705] "Dynamically optimized food and drink" refers to food and drink suggestions that are tailored based on the user's emotions and physical condition.

[1706] "Generative AI model" refers to an artificial intelligence model used to perform a generative task.

[1707] "Emergency Support" means an emergency response required immediately by a User.

[1708] An "alert" is a notification that notifies you of an emergency or important information.

[1709] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. The system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions.

[1710] System configuration

[1711] This system includes a user device, a server, a database, a natural language processing engine, an emotion engine, and a machine learning model. The user device refers to an electronic device such as a smartphone, tablet, or PC, and the server is a computer system for receiving, processing, and transmitting data.

[1712] Program processing

[1713] Login and Access

[1714] A user accesses the system's website or application using their own device and enters their username and password to log in. This authentication information is sent to the server, and once the server has successfully authenticated, the user is provided with a chat interface for consultation.

[1715] Enter and submit your inquiry

[1716] The user enters the content of the consultation in text format into the chat interface, and this content is sent to the server.

[1717] Initial analysis and sentiment analysis

[1718] The server uses natural language processing technology to perform an initial analysis of the received consultation content, and then calculates an emotion score using an emotion engine, which allows the user's emotional state to be numerically evaluated.

[1719] Topic classification and historical data reference

[1720] Based on the analysis results, the consultation content is classified into topics, and similar consultation content and solutions are extracted by referring to a database of past consultation content.

[1721] Proposal generation and optimization

[1722] The server generates recommendations based on the extracted solutions and dynamically optimizes them based on the emotion score, suggesting foods and drinks that correspond to the user's emotions and physical condition.

[1723] View suggestions and get feedback

[1724] The generated suggestions are sent back to the user's device and displayed to the user. The user checks the suggestions and enters feedback. This feedback is sent to the server and stored in a database.

[1725] Training a machine learning model

[1726] The server uses the accumulated feedback to train a machine learning model and generate more accurate support for the next consultation.

[1727] Hardware and software used

[1728] Hardware: User devices (smartphones, tablets, PCs), servers (high-performance computing systems)

[1729] Software: Natural language processing engines (e.g., SpaCy, NLTK), emotion engines (e.g., Hugging Face's transformers library), machine learning models (e.g., Scikit-learn, TensorFlow)

[1730] Specific examples

[1731] Example 1:

[1732] If a user types, "I've been stressed out lately and have lost my appetite.":

[1733] The system suggests, "Would you like some relaxing herbal tea?"

[1734] The user orders the suggested herbal tea and enters feedback as "That was a good suggestion, I feel better."

[1735] Prompt Sentence Examples

[1736] "The user has recently been feeling stressed and has lost their appetite. Please suggest suitable foods and drinks. Use your historical database and emotion engine to generate the best suggestions."

[1737] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1738] Specific processing flow of the program

[1739] Step 1: User Login and Access

[1740] Input: The user enters their username and password from their device (smartphone, tablet, computer).

[1741] Data processing: The device sends the entered authentication information (username and password) to the server.

[1742] Data calculation: The server checks the authentication information against the database.

[1743] Output: If authentication is successful, the server starts a session and presents the user with a chat interface for consultation.

[1744] Step 2: Enter and submit your inquiry

[1745] Input: The user enters the content of their inquiry in text format into the chat interface.

[1746] Data processing: The device sends the consultation details entered to the server.

[1747] Data calculation: None (simple data transmission).

[1748] Output: The consultation content is received by the server.

[1749] Step 3: Initial analysis and sentiment analysis

[1750] Input: Text data of the consultation received by the server.

[1751] Data processing: The server performs initial analysis of the text data using a natural language processing engine (e.g., SpaCy, NLTK).

[1752] Data calculation: An emotion engine (e.g., Hugging Face transformers library) is used to calculate an emotion score and numerically evaluate the emotional state of the consultation.

[1753] Output: Analysis results of the text data and sentiment scores.

[1754] Step 4: Topic classification and historical data reference

[1755] Input: Initial analysis results and sentiment scores.

[1756] Data processing: The server categorizes the consultation content into topics based on the analysis results.

[1757] Data calculation: Query the database of past consultations to extract similar consultations and their solutions.

[1758] Output: Topic classification results, and similar consultations and their solutions.

[1759] Step 5: Generate and optimize proposals

[1760] Input: Similar consultations and their solutions, sentiment scores.

[1761] Data processing: The server generates suggestions using the generative AI model.

[1762] Data computation: Dynamically optimizes recommendations based on emotion scores, suggesting foods and drinks that correspond to the user's emotions and physical condition.

[1763] Output: Dynamically optimized suggestions.

[1764] Step 6: View your proposal and get feedback

[1765] Input: Dynamically optimized suggestions.

[1766] Data processing: The server sends the proposal to the user's device.

[1767] Data calculation: None (simple data transmission).

[1768] Output: The suggestion is displayed on the user's device.

[1769] Step 7: Capture and store feedback

[1770] Input: User feedback.

[1771] Data processing: The device sends feedback to the server.

[1772] Data calculation: The server stores the received feedback in a database.

[1773] Output: Feedback stored in a database.

[1774] Step 8: Train the machine learning model

[1775] Input: Accumulated feedback data.

[1776] Data processing: The server inputs the feedback data into the machine learning model.

[1777] Data Computing: Using machine learning models (e.g., Scikit-learn, TensorFlow) to train the next consultation to improve its accuracy.

[1778] Output: An improved machine learning model.

[1779] Specific examples

[1780] Example 1: If a user enters, "I've been feeling stressed lately and have no appetite," the system will suggest, "How about some herbal tea to help you relax?" and the user will confirm and enter their feedback.

[1781] Example prompt sentence:

[1782] "The user has recently been feeling stressed and has lost their appetite. Please suggest suitable foods and drinks. Use your historical database and emotion engine to generate the best suggestions."

[1783] 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.

[1784] 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.

[1785] 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.

[1786] [Fourth embodiment]

[1787] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1788] 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.

[1789] 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).

[1790] 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.

[1791] 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.

[1792] 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).

[1793] 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.

[1794] 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.

[1795] 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.

[1796] 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.

[1797] 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.

[1798] 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.

[1799] 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."

[1800] This invention provides a system that allows users to receive mental health consultations and appropriate support without having to deal with face-to-face consultations. The system is composed of multiple means. Each means and its specific processing will be described below.

[1801] User Access and Login

[1802] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[1803] Enter and submit your inquiry

[1804] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[1805] Receiving consultation content and initial analysis

[1806] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, it performs sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[1807] Historical database queries

[1808] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[1809] Proposal generation and delivery

[1810] The server sends the generated suggestions to the device, which displays them in the chat interface and allows the user to review the advice or solution provided.

[1811] Receive feedback and learn

[1812] The user inputs feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation.

[1813] Specific examples

[1814] Example 1: Work stress

[1815] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[1816] The terminal transmits the consultation contents to the server.

[1817] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigning a low emotional score and classifying the topic as "work stress."

[1818] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[1819] The server generates the proposal and sends it to the device.

[1820] The device displays the suggestions to the user.

[1821] Users try out the proposed methods and provide feedback.

[1822] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[1823] Example 2: Relationship problems

[1824] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[1825] The terminal transmits the consultation contents to the server.

[1826] The server analyzes text data containing the keywords "colleague" and "poor relationship" and classifies sentiment scores and topics.

[1827] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[1828] The server generates the proposal and sends it to the device.

[1829] The device displays the suggestions to the user.

[1830] The user reviews the solution and provides feedback.

[1831] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[1832] This system provides users with an environment where they can seek advice anonymously and with peace of mind, and continuous feedback improves the accuracy of the system.

[1833] The processing flow will be explained below.

[1834] Step 1:

[1835] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[1836] Step 2:

[1837] The device displays a login screen and the user enters their username and password.

[1838] Step 3:

[1839] The terminal transmits the entered authentication information to the server.

[1840] Step 4:

[1841] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[1842] Step 5:

[1843] The server displays a dashboard or chat interface to the authenticated user.

[1844] Step 6:

[1845] The user enters the content of the consultation in text format into the chat interface.

[1846] Step 7:

[1847] The terminal transmits the input consultation content to the server.

[1848] Step 8:

[1849] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[1850] Step 9:

[1851] As part of the initial analysis, the server performs sentiment analysis to calculate a sentiment score for the consultation content, and also performs topic classification to assign the content to the appropriate category.

[1852] Step 10:

[1853] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[1854] Step 11:

[1855] The server organizes the extracted solutions and generates proposals.

[1856] Step 12:

[1857] The server sends the generated proposal to the terminal.

[1858] Step 13:

[1859] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[1860] Step 14:

[1861] The user enters feedback on the provided suggestions into a chat interface.

[1862] Step 15:

[1863] The device sends the user's feedback to the server.

[1864] Step 16:

[1865] The server receives the feedback and stores it in a database.

[1866] Step 17:

[1867] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[1868] Step 18:

[1869] The next time the user uses the system, the server will use the updated machine learning model to provide more personalized support to the user.

[1870] Example 1

[1871] 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."

[1872] Conventional mental health consultation systems have difficulty providing appropriate support to users who wish to avoid in-person consultations. Furthermore, these systems lack sufficient mechanisms for effectively utilizing user feedback to improve the accuracy of subsequent consultations. Furthermore, they often lack the means to properly analyze the content of the consultation and provide appropriate solutions. To address these issues, the present invention provides a system that allows users to safely seek mental health consultations and receive appropriate support while avoiding in-person consultations.

[1873] 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.

[1874] In this invention, the server includes: means for a user to input a consultation content in text format; means for a terminal to transmit the consultation content to the server; means for the server to receive the consultation content and perform an initial analysis using natural language processing technology; means for classifying the consultation content by topic based on the results of the initial analysis; means for querying a database of past consultation content and extracting similar consultation content and solutions; means for organizing the extracted solutions and generating a proposal content to be provided to the user; means for transmitting the generated proposal content to the user's terminal and displaying it to the user; means for receiving feedback from the user and storing the feedback in a database; and means for training a machine learning model based on the stored data to generate more accurate support for the next consultation. This allows users to receive mental health consultation anonymously and with peace of mind, and by utilizing continuous feedback, the system can provide more accurate support for subsequent consultations.

[1875] "User" refers to a person who uses the system to seek mental health advice.

[1876] "Terminal" refers to a device, such as a personal computer or smartphone, that a user uses to access the system and enter consultation details.

[1877] "Server" refers to a computer system that processes the inquiry content sent by the user and generates and provides proposal content.

[1878] "Consultation content" refers to text information entered by users about questions and concerns about mental health.

[1879] "Natural language processing technology" refers to technology that analyzes text data, calculates sentiment scores, and performs topic classification.

[1880] "Initial analysis" refers to the initial analysis process that the server performs after receiving the consultation content using natural language processing technology.

[1881] "Topic classification" refers to the process of categorizing consultation content into specific categories or themes based on the results of the initial analysis.

[1882] A "database" refers to a collection of information that stores and manages past consultation details and their solutions.

[1883] A "machine learning model" refers to an algorithm or system that is trained to provide more accurate support for the next consultation based on feedback accumulated in a database.

[1884] "Sentiment analysis" refers to the technology of calculating emotional trends and emotional scores from text data of consultation content.

[1885] "Emotion score" refers to the numerical representation of the intensity and tendency of emotions contained in the consultation content as a result of emotion analysis.

[1886] "Proposal" refers to advice or solutions generated by the server and provided to the user.

[1887] "Feedback" refers to the ratings and comments that users enter in response to suggestions.

[1888] This invention relates to a system that allows users to receive mental health consultations and appropriate support in a safe environment while avoiding face-to-face consultations. The specific configuration and operation of this system will be described below.

[1889] User Access and Login

[1890] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen, and the user enters a username and password. The entered authentication information is sent from the device to the server. The server verifies the authentication information, and if authentication is successful, starts a session and displays a dashboard and chat interface to the user.

[1891] Enter and submit your inquiry

[1892] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[1893] Receiving consultation content and initial analysis

[1894] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology (e.g., Google Cloud Natural Language API, AWS Comprehend, etc.). The analysis includes sentiment analysis and topic classification. Sentiment analysis calculates an emotional score for the consultation content, and topic classification classifies the consultation content into specific categories.

[1895] Historical database queries

[1896] Based on the initial analysis results, the server queries a database of past consultations (e.g., MySQL, PostgreSQL, etc.) to extract similar consultations and their solutions. The server then organizes these extracted solutions and generates proposals to be provided to users.

[1897] Proposal generation and delivery

[1898] The server generates suggestions and sends them to the device, which displays them in a chat interface, allowing the user to review the advice and solutions provided.

[1899] Receive feedback and learn

[1900] The user enters feedback on the provided suggestions, and the device sends the user's feedback to the server, which stores the feedback in a database and trains a machine learning model to provide more accurate support for the next consultation.

[1901] Specific examples

[1902] Example 1: Work stress

[1903] 1. A user says, "I haven't been able to sleep lately because of work stress."

[1904] 2. The device sends the consultation details to the server.

[1905] 3. The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," assigns a low emotional score, and classifies the topic as "work stress."

[1906] 4. The server queries the historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[1907] 5. The server generates the proposal and sends it to the device.

[1908] 6. The device displays the suggestions to the user.

[1909] 7. The user tries out the proposed solution and provides feedback.

[1910] 8. The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[1911] Example 2: Relationship problems

[1912] 1. A user comes to us with a problem: "I'm having trouble with my coworkers."

[1913] 2. The device sends the consultation details to the server.

[1914] 3. The server analyzes text data containing the keywords "colleague" and "relationship not going well" and classifies sentiment scores and topics.

[1915] 4. The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[1916] 5. The server generates the proposal and sends it to the device.

[1917] 6. The device displays the suggestions to the user.

[1918] 7. The user reviews the solution and provides feedback.

[1919] 8. The device sends the feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[1920] Prompt Sentence Examples

[1921] "I haven't been able to sleep lately because of work stress. What should I do?"

[1922] "I'm having trouble with my coworkers. How can I improve it?"

[1923] In this way, users can seek mental health advice anonymously and with peace of mind, and by utilizing continuous feedback, the system will be able to provide more accurate support for subsequent consultations.

[1924] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1925] Step 1:

[1926] A user accesses the system's website or application using a device such as a personal computer or smartphone. The input is a username and password, and the device displays a login screen, which collects the authentication information. The output is the authentication information entered by the user.

[1927] Step 2:

[1928] The device sends the entered authentication information to the server. The input is the username and password entered by the user, and the output is the authentication information sent to the server. The device securely transmits the authentication information using the SSL / TLS protocol.

[1929] Step 3:

[1930] The server queries the user database based on the received authentication information to verify authentication. The input is the received authentication information, and the output is whether the authentication was successful. If authentication is successful, the server generates a session ID and generates information to display the dashboard and chat interface.

[1931] Step 4:

[1932] The user inputs the content of the consultation in text format into the chat interface. The input is the text of the consultation content, and the output is the input text data.

[1933] Step 5:

[1934] The terminal transmits the input consultation content to the server. The input is text data entered by the user, and the output is text data transmitted to the server. The terminal transmits the data using a communication protocol.

[1935] Step 6:

[1936] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. The input is the received text data, and the output is the analysis results. The server performs sentiment analysis, calculates sentiment scores, and performs topic classification.

[1937] Step 7:

[1938] The server queries a database of past consultations based on the initial analysis results. The input is the initial analysis results, and the output is similar consultations and their solutions. The server queries the database and organizes the extracted solutions.

[1939] Step 8:

[1940] The server generates a proposal to be provided to the user based on the extracted solution. The input is the extracted solution, and the output is the proposal. The server sends the generated proposal to the terminal.

[1941] Step 9:

[1942] The terminal displays the suggestions in a chat interface, and the user confirms the offered advice or solution. The input is the suggestion sent from the server, and the output is the suggestion displayed in the chat interface.

[1943] Step 10:

[1944] The user inputs feedback on the provided suggestion. The input is the feedback, and the output is the input feedback.

[1945] Step 11:

[1946] The terminal sends the user's feedback to the server. The input is the feedback content, and the output is the feedback sent to the server. The terminal transmits the data using a communication protocol.

[1947] Step 12:

[1948] The server accumulates the feedback in a database and trains a machine learning model. The input is the feedback content, and the output is the trained machine learning model. This makes it possible to provide more accurate support for the next consultation.

[1949] (Application example 1)

[1950] 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."

[1951] Mental health care for drivers in autonomous vehicles is an important issue for ensuring safe driving and reducing driver stress. However, conventional systems have no way for drivers to receive appropriate support in real time while avoiding face-to-face consultations. This has led to a lack of mental health follow-up in emergencies while driving.

[1952] 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.

[1953] In this invention, the server includes means for a user to input a consultation content in text format, means for receiving the consultation content in the server and performing an initial analysis using natural language processing, means for topic categorizing the consultation content based on the results of the initial analysis, means for consulting a database of past consultation content and extracting similar consultation contents and their solutions, means for organizing the extracted solutions and generating a proposal content to be provided to the user, means for sending the generated proposal content to the user's terminal and displaying it to the user, means for receiving feedback from the user and storing the feedback in a database, means for training a machine learning model based on the stored data and generating more accurate support for the next consultation, means for providing an interface for providing mental health care in an autonomous vehicle, and means for monitoring the driver's mental state while driving and providing support in real time, thereby enabling the driver to receive mental health care in real time in an autonomous vehicle.

[1954] "User" refers to a driver who uses the system to seek mental health advice.

[1955] "Text consultation content" refers to the string of characters that users enter into the system, and the content represents mental health issues and concerns.

[1956] The "server" is the computer at the core of the system, which receives inquiries from users and analyzes them and provides support suggestions.

[1957] "Natural language processing" is a technology that allows a computer to analyze text data entered by a user and understand its meaning and emotions.

[1958] "Initial analysis" is the process of first analyzing the consultation content using natural language processing technology and extracting emotion scores and keywords.

[1959] "Topic classification" refers to classifying the content of a consultation into multiple categories based on the results of the initial analysis.

[1960] The "past consultation content database" is a database that stores consultation content and solutions that have been accumulated in the system up to now.

[1961] "Similar consultation content" refers to cases in the database of past consultation content that are similar in content and emotion to the current consultation content.

[1962] "Solutions" are advice or measures provided to resolve a user's mental health issues based on past consultations.

[1963] The "proposal content" is a support proposal generated for the current consultation content based on the extracted solution.

[1964] "User's device" means a device used by a user, including a smartphone, tablet, or in-vehicle interface.

[1965] "Feedback" refers to the opinions and evaluations that users give in response to suggestions, and is used to improve the accuracy of the system.

[1966] A "machine learning model" is an algorithm that learns from large amounts of data and generates more accurate support for the next consultation.

[1967] An "autonomous vehicle" is a vehicle that automatically performs driving operations and provides a place for the driver to receive mental health care.

[1968] An "interface" is an operating means, such as a screen or voice input system, that allows a user to interact with a system.

[1969] "Mental state monitoring" is the process of monitoring a driver's emotions and stress levels and collecting data in real time.

[1970] This invention is a system that allows users to receive mental health consultations and appropriate support while in an autonomous vehicle. The specific configuration and procedure are shown below.

[1971] 1. System Configuration

[1972] Main components:

[1973] User devices: devices such as touchscreens in vehicles and smartphones

[1974] Server: Cloud-based computing cluster

[1975] Natural language processing models, such as BERT for Hugging Face

[1976] Database: A database for storing past consultations and feedback

[1977] 2. User Access and Login

[1978] Users access the system using a touchscreen in their vehicle or a smartphone, enter their username and password, and the device sends the credentials to a server. After authentication, the session begins and the user is presented with a dashboard and chat interface.

[1979] 3. Enter and submit your inquiry

[1980] The user inputs the text of the consultation into the chat interface, and the device sends the inputted consultation content to the server.

[1981] 4. Receiving consultation content and initial analysis

[1982] When the server receives the consultation content, it performs an initial analysis of the text data using a natural language processing model. As part of the analysis, it performs a sentiment analysis and calculates an emotion score for the consultation content. It also categorizes the consultation content into topics and assigns them to appropriate categories.

[1983] 5. Query the database and generate suggestions

[1984] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions, and then organizes the extracted solutions to generate proposals to be provided to the user.

[1985] 6. Providing suggestions and feedback

[1986] The server generates suggestions and sends them to the user's device, which displays them in the chat interface. The user reviews the advice and solutions provided and enters feedback on the results. The device then sends the feedback to the server, which stores it in a database.

[1987] 7. Feedback learning and model training

[1988] The server uses the accumulated feedback to train a machine learning model and provide more accurate support for the next consultation.

[1989] 8. Operation in autonomous vehicles

[1990] The system can monitor the driver's mental state while driving in an autonomous vehicle and provide appropriate support in real time. For example, if the driver feels anxious on the highway, the system can suggest deep breathing or meditation to relieve tension.

[1991] Specific examples

[1992] 1. Example 1: Highway anxiety

[1993] Consultation: "Recently, I've been feeling anxious when driving on the highway."

[1994] Suggestion: "To reduce highway anxiety, try listening to appropriate music and taking deep breaths."

[1995] Feedback: "I tried the deep breathing exercises you suggested and my anxiety has decreased a little."

[1996] 2. Example 2: Commuting stress

[1997] Consultation: "I always feel stressed during my morning commute."

[1998] Suggestion: "Trying a short meditation session before you start driving may help reduce stress."

[1999] Feedback: "I tried meditation and it made me feel a little less stressed in the mornings."

[2000] In this way, the driver can receive real-time mental health care in the vehicle, providing a safe and comfortable driving environment.

[2001] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2002] Step 1:

[2003] The user enters their username and password using the vehicle's touchscreen or smartphone.

[2004] Input: Username, Password

[2005] Output: Credentials

[2006] Specific operation:

[2007] The user operates the login screen on the vehicle's touchscreen and enters their username and password.

[2008] Step 2:

[2009] The terminal transmits the entered authentication information to the server, and the server performs authentication.

[2010] Input: Credentials

[2011] Output: Authentication result (success / failure)

[2012] Specific operation:

[2013] The device sends the user's input information to the server via encrypted communication such as SSL, and the server compares it with a database and returns the authentication result.

[2014] Step 3:

[2015] If authentication is successful, the server starts a session and presents the user with a dashboard and chat interface.

[2016] Input: Authentication result (success)

[2017] Output: Dashboard, chat interface

[2018] Specific operation:

[2019] The server generates a session ID and sends it to the user's device, which receives it and displays the dashboard and chat interface.

[2020] Step 4:

[2021] The user inputs the content of the consultation in text format into the chat interface, and the terminal transmits the content to the server.

[2022] Input: Consultation content (text format)

[2023] Output: Sending the consultation details to the server

[2024] Specific operation:

[2025] A user types "Lately, I feel anxious when I drive on the highway" into the chat interface and clicks the send button. The device then sends this text to the server.

[2026] Step 5:

[2027] The server receives the consultation content and performs an initial analysis using a natural language processing model.

[2028] Input: Consultation content (text format)

[2029] Output: Initial analysis results (sentiment score, topic classification)

[2030] Specific operation:

[2031] The server performs sentiment analysis using Hugging Face's BERT model and other tools to analyze the content of the consultation. The analysis results include "anxiety (NEGATIVE)."

[2032] Step 6:

[2033] The server queries a database of past consultation contents based on the results of the initial analysis and extracts similar consultation contents and their solutions.

[2034] Input: Initial analysis results (sentiment score, topic classification)

[2035] Output: Similar inquiries and solutions

[2036] Specific operation:

[2037] The server uses the analysis results to query a database and extract similar consultations and past solutions (e.g., deep breathing, meditation, etc.).

[2038] Step 7:

[2039] The server organizes the extracted solutions and generates proposals to be presented to the user.

[2040] Input: Similar inquiries and solutions

[2041] Output: Proposal

[2042] Specific operation:

[2043] The server organizes solutions and generates suggestions such as, "To reduce anxiety on the highway, try listening to appropriate music or taking deep breaths."

[2044] Step 8:

[2045] The server sends the generated proposals to the user's device, which then displays them to the user.

[2046] Input: Proposal

[2047] Output: what is displayed to the user

[2048] Specific operation:

[2049] The server transmits the proposal to the terminal, which displays the proposal on the chat interface.

[2050] Step 9:

[2051] The user tries the proposed method and inputs feedback on the results, which is then sent to the server.

[2052] Input: Feedback (text format)

[2053] Output: Send feedback

[2054] Specific operation:

[2055] The user types, "I tried the suggested deep breathing exercises and my anxiety has decreased a little," and submits the feedback. The device then sends this feedback to the server.

[2056] Step 10:

[2057] The server receives the feedback, stores it in a database, and trains a machine learning model to provide more accurate support for the next consultation.

[2058] Input: Feedback

[2059] Output: A trained machine learning model

[2060] Specific operation:

[2061] The server stores the feedback in a database and uses this data to update and train the machine learning model, improving the accuracy of support for the next consultation.

[2062] 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.

[2063] The present invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing are explained below.

[2064] User Access and Login

[2065] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[2066] Enter and submit your inquiry

[2067] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[2068] Receiving consultation content and initial analysis

[2069] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The server also categorizes the consultation content into topics and assigns them to appropriate categories.

[2070] Historical database queries

[2071] The server queries a database of past consultations based on the initial analysis results to extract similar consultations and their solutions, organizes the extracted solutions, and generates proposals to provide to the user.

[2072] Emotional engine regulation

[2073] The server uses an emotion engine to dynamically adjust its suggestions based on the user's emotion score, providing advice and solutions that are optimized for the user's emotions.

[2074] Proposal generation and delivery

[2075] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[2076] Receive feedback and learn

[2077] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[2078] Specific examples

[2079] Example 1: Work stress

[2080] A user complains, "I haven't been able to sleep lately because of work stress."

[2081] The terminal transmits the consultation contents to the server.

[2082] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[2083] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[2084] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[2085] The server sends the proposal to the device.

[2086] The device displays the suggestions to the user.

[2087] Users try out the proposed methods and provide feedback.

[2088] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[2089] Example 2: Relationship problems

[2090] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[2091] The terminal transmits the consultation contents to the server.

[2092] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[2093] The server queries a historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[2094] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[2095] The server sends the proposal to the device.

[2096] The device displays the suggestions to the user.

[2097] The user implements the proposed method and provides feedback.

[2098] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[2099] This system provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, it provides personalized support optimized to the user's emotions.Continuous feedback constantly improves the system's accuracy, resulting in more effective support.

[2100] The processing flow will be explained below.

[2101] Step 1:

[2102] A user accesses the system's website or application using a device such as a personal computer or smartphone.

[2103] Step 2:

[2104] The device displays a login screen and the user enters their username and password.

[2105] Step 3:

[2106] The terminal transmits the entered authentication information to the server.

[2107] Step 4:

[2108] The server verifies the authentication information and authenticates the user. If authentication is successful, the session begins.

[2109] Step 5:

[2110] The server displays a dashboard or chat interface to the authenticated user.

[2111] Step 6:

[2112] The user enters the content of the consultation in text format into the chat interface.

[2113] Step 7:

[2114] The terminal transmits the input consultation content to the server.

[2115] Step 8:

[2116] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[2117] Step 9:

[2118] As part of the initial analysis, the server uses an emotion engine to calculate an emotion score and recognize the emotion of the consultation content.

[2119] Step 10:

[2120] The server classifies the consultation topic and assigns it to the appropriate category.

[2121] Step 11:

[2122] The server queries a database of past consultations based on the initial analysis results and extracts similar consultations and their solutions.

[2123] Step 12:

[2124] The server organizes the extracted solutions and dynamically adjusts the proposals based on the emotion score using an emotion engine.

[2125] Step 13:

[2126] The server sends the generated proposal to the terminal.

[2127] Step 14:

[2128] The device displays the suggestions in a chat interface, allowing the user to review the advice and solutions.

[2129] Step 15:

[2130] The user enters feedback on the provided suggestions into a chat interface.

[2131] Step 16:

[2132] The device sends the user's feedback to the server.

[2133] Step 17:

[2134] The server receives the feedback and stores it in a database.

[2135] Step 18:

[2136] The server trains a machine learning model based on the accumulated data and generates more accurate support for the next consultation.

[2137] Step 19:

[2138] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[2139] Step 20:

[2140] The next time the system is used, the server will use the updated machine learning model and emotional data to provide more accurate, optimized support to the user.

[2141] Example 2

[2142] 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."

[2143] In recent years, mental health issues have been increasing, creating a demand for effective and prompt support. However, it is difficult to provide an anonymous environment where users can seek advice in a safe and secure environment. Furthermore, there are limited systems that provide appropriate advice based on the user's emotions and the content of their consultation. As a result, the accuracy and effectiveness of responses to users' consultations are often insufficient.

[2144] 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.

[2145] In this invention, the server includes a means for a user to input the consultation content in text format, a means for receiving the consultation content in the server and performing an initial analysis using natural language processing, and a means for dynamically adjusting the content of suggestions based on the user's emotion score. This provides an environment where users can consult anonymously with peace of mind, and by using an emotion engine, it becomes possible to provide personalized support optimized for the user's emotions in real time.

[2146] "User" refers to an individual who accesses the system and inputs the details of their consultation.

[2147] "Terminal" refers to a device used by a user, such as a personal computer or smartphone.

[2148] "Server" refers to a central computer system that receives and analyzes consultation content, generates proposals, and provides them to users.

[2149] "Natural language processing" refers to the technology for analyzing input text data and understanding its meaning and emotions.

[2150] "Initial analysis" refers to the process of first analyzing the text of the input consultation content, calculating the emotion score, and classifying topics.

[2151] An "emotion engine" refers to an algorithm or model that calculates an emotional score from text data entered by the user and recognizes their emotional state.

[2152] "Emotion score" refers to a numerical value that indicates the user's emotional state, calculated based on the input text data.

[2153] "Topic classification" refers to the process of sorting the input consultation content into appropriate categories based on its content.

[2154] "Database" refers to a storage device or system for storing past consultation details and feedback data.

[2155] A "machine learning model" refers to an algorithm or program that is trained using collected data to provide more accurate support for the next consultation.

[2156] "Suggestion" refers to advice or solutions generated by the server and provided to the user.

[2157] "Feedback" refers to the opinions and evaluations that users enter regarding the suggestions.

[2158] "Dynamic adjustment" refers to changing the suggestions in real time in response to changes in the user's emotional score, etc.

[2159] "Emergency Support" refers to fast, special support provided when a user's inquiry requires urgent attention.

[2160] "Alert" means a warning or notification generated when it is determined that emergency support is required.

[2161] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. This system is characterized by incorporating an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions. Each of these means and their specific processing will be explained below.

[2162] User Access and Login

[2163] A user accesses the system's website or application using a device such as a personal computer or smartphone. The device displays a login screen where the user enters their username and password. The device sends the entered authentication information to the server, which verifies it. If authentication is successful, the server starts a session and displays a dashboard or chat interface to the user.

[2164] Enter and submit your inquiry

[2165] The user inputs the content of the consultation in text format into the chat interface, and the device sends the input content to the server.

[2166] Receiving consultation content and initial analysis

[2167] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology. As part of the analysis, an emotion engine is used to calculate an emotion score and recognize the emotion of the consultation content. The consultation content is also classified into topics and sorted into appropriate categories. Natural language processing technology used includes Python's NLTK and SpaCy.

[2168] Historical database queries

[2169] The server queries a database of past consultations based on the results of the initial analysis to extract similar consultations and their solutions. It then organizes the extracted solutions and generates proposals to be provided to users. To achieve this, a database management system (e.g., MySQL or PostgreSQL) is used.

[2170] Emotional engine regulation

[2171] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score, providing advice and solutions optimized for the user's emotions. The emotion engine uses emotion analysis models such as BERT and RoBERTa.

[2172] Proposal generation and delivery

[2173] The server generates suggestions and sends them to the device, which displays them in a chat interface for the user to review and confirm the advice or solution.

[2174] Receive feedback and learn

[2175] Users enter their feedback on the suggestions provided into the chat interface. The device then sends the user's feedback to the server, which stores it in a database and trains a machine learning model to provide more accurate support for the next consultation. The emotion engine also tracks the user's long-term emotional changes and provides personalized support.

[2176] Specific examples

[2177] Example 1: Work stress

[2178] A user comes to us with a complaint, "I haven't been able to sleep lately because of work stress."

[2179] The terminal transmits the consultation contents to the server.

[2180] The server uses natural language processing technology to analyze text data containing the keywords "stress" and "can't sleep," and then uses an emotion engine to assign a low emotion score and classify the topic as "work stress."

[2181] The server queries a historical database to extract similar cases and their solutions (e.g., meditation, exercise, time management techniques).

[2182] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[2183] The server sends the proposal to the device.

[2184] The device displays the suggestions to the user.

[2185] Users try out the proposed methods and provide feedback.

[2186] The device sends feedback to the server, which stores it in a database to improve accuracy in subsequent consultations.

[2187] Example 2: Relationship problems

[2188] A user comes to us with a problem: "I'm having trouble getting along with my coworkers."

[2189] The terminal transmits the consultation contents to the server.

[2190] The server analyzes text data containing the keywords "colleague" and "poor relationship" and uses an emotion engine to classify emotion scores and topics.

[2191] The server queries the historical database and extracts solutions to workplace communication problems (e.g., question and answer techniques, ways to improve communication skills).

[2192] The server generates the suggestions and uses an emotion engine to dynamically adjust the suggestions based on the emotion scores.

[2193] The server sends the proposal to the device.

[2194] The device displays the suggestions to the user.

[2195] The user implements the proposed method and provides feedback.

[2196] The device sends feedback information to the server, which updates the database and aims to improve the accuracy of the next consultation.

[2197] This invention provides an environment where users can seek advice anonymously and with peace of mind, and by using an emotion engine, personalized support optimized for the user's emotions is provided in real time. Continuous feedback constantly improves the accuracy and effectiveness of the system.

[2198] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2199] Step 1:

[2200] A user accesses the system's website or application using a terminal such as a personal computer or smartphone.

[2201] Input: Website URL or application icon

[2202] Output: Login screen displayed

[2203] What happens: The user enters a URL in a web browser or taps an application to launch it. The device renders and displays a login screen to the user.

[2204] Step 2:

[2205] The device will display a login screen and prompt you to enter your username and password.

[2206] Input: None

[2207] Output: Username and password input fields displayed

[2208] What it does: The terminal renders a login form with username and password input fields.

[2209] Step 3:

[2210] The user enters their username and password and clicks the submit button.

[2211] Input: Username and Password

[2212] Output: Send authentication information to the server

[2213] How it works: The user enters information on the keyboard and clicks the login button. The device sends the information to the server.

[2214] Step 4:

[2215] Verify the authentication information received by the server.

[2216] Input: Username and Password

[2217] Output: Authentication result

[2218] What it does: The server queries its database to verify that the username and password submitted match.

[2219] Step 5:

[2220] If authentication is successful, the server starts a session and presents the user with a dashboard or chat interface.

[2221] Input: Authentication result (success)

[2222] Output: Display URL for dashboard or chat interface

[2223] How it works: The server generates a session ID, saves the session data, and responds with a URL for the dashboard or chat interface, which the device receives and displays to the user.

[2224] Step 6:

[2225] The user enters the content of the consultation in text format into the chat interface.

[2226] Input: Text of consultation

[2227] Output: Consultation content data

[2228] Action: The user enters text using a keyboard or touchscreen and clicks the send button. The device sends the entered text data to the server.

[2229] Step 7:

[2230] The server receives the consultation content and performs an initial analysis of the text data using natural language processing technology.

[2231] Input: Consultation content data

[2232] Output: Initial analysis results (tokenized text and sentiment scores)

[2233] How it works: The server uses Python libraries such as NLTK and SpaCy to tokenize and morphologically analyze the text and calculate a sentiment score.

[2234] Step 8:

[2235] The server categorizes the consultation content into topics based on the initial analysis results.

[2236] Input: Initial analysis results

[2237] Output: Topic classification results

[2238] How it works: The server uses topic models (LDA or LSI) to classify text data into categories.

[2239] Step 9:

[2240] The server queries a database of past consultations and extracts similar consultations and their solutions.

[2241] Input: Topic classification results

[2242] Output: Similar issues and solutions

[2243] How it works: The server generates an SQL query and sends it to the database to retrieve similar queries and their solutions.

[2244] Step 10:

[2245] The server uses an emotion engine to dynamically adjust the suggestions based on the user's emotion score.

[2246] Input: Similar issues and solutions, sentiment scores

[2247] Output: Dynamically adjusted suggestions

[2248] How it works: The server adjusts the tone and content of the solution it presents based on the sentiment score.

[2249] Step 11:

[2250] The server sends the generated proposal to the terminal.

[2251] Input: Dynamically adjusted suggestions

[2252] Output: Proposal content data

[2253] How it works: The server sends the suggestions as an HTTP response, which the device receives and renders in the chat interface.

[2254] Step 12:

[2255] Users enter feedback on the suggestions provided into a chat interface.

[2256] Input: Feedback text

[2257] Output: Feedback data

[2258] Action: The user enters feedback in text format and clicks the send button. The device sends the feedback data to the server.

[2259] Step 13:

[2260] The server accumulates user feedback in a database and trains a machine learning model.

[2261] Input: Feedback data

[2262] Output: An updated machine learning model

[2263] How it works: The server stores the feedback data in a database and periodically retrains the machine learning model.

[2264] Step 14:

[2265] The server uses an emotion engine to track the user's long-term emotional changes and provide personalized support.

[2266] Input: User's historical sentiment score

[2267] Output: Personalized support

[2268] How it works: The server analyzes historical data of sentiment scores and applies an algorithm to track long-term sentiment trends.

[2269] (Application example 2)

[2270] 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."

[2271] While existing mental health support systems provide an environment where users can seek advice anonymously and with peace of mind, they do not adequately respond appropriately to the user's emotions. Furthermore, they lacked mechanisms for effectively utilizing user feedback to improve the accuracy of future consultations, making it difficult to provide optimal support tailored to each user's individual situation. Furthermore, there were no systems that could suggest foods or beverages based on the content of mental health consultations, meaning there was a lack of a means to provide immediate, specific support for specific emotions or physical conditions.

[2272] 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.

[2273] In this invention, the server includes: a means for a user to input a text-based consultation content; a means for receiving the consultation content in the server and performing an initial analysis using natural language processing; a means for topic-classifying the consultation content based on the results of the initial analysis; a means for querying a database of past consultation content to extract similar consultation content and solutions; a means for organizing the extracted solutions and generating a proposal content to be provided to the user; a means for sending the generated proposal content to the user's terminal and displaying it to the user; a means for receiving feedback from the user and storing the feedback in a database; a means for training a machine learning model based on the accumulated data to generate more accurate support for the next consultation; a means for dynamically optimizing food and beverage suggestions based on the user's consultation content and emotion score; a means for generating food and beverage suggestions based on the user's emotion using a generative AI model; and a means for generating an alert if the generated proposal content determines that emergency support is required. This allows for the provision of proposal content optimized to the user's emotion, enabling immediate food and beverage suggestions tailored to specific emotions or physical conditions. Furthermore, by utilizing user feedback, the system's accuracy can be improved, enabling more personalized support to be provided.

[2274] "User" refers to an individual who uses the System.

[2275] "Text-format consultation content" refers to text information that a user inputs into the system, including consultations and questions.

[2276] "Server" refers to a computing system that receives, processes, and transmits data.

[2277] "Natural language processing" is a technology that uses computers to understand, analyze, and generate human language.

[2278] "Initial analysis" is the process of first analyzing the input text data to understand its basic structure and meaning.

[2279] "Topic classification" refers to the process of sorting text data into specific subjects or categories.

[2280] "Past consultation content database" refers to a data bank that stores consultation content and solutions that users have previously entered into the system.

[2281] "Similar consultation contents and their solutions" refers to consultation contents and their solutions that are highly similar to the current consultation contents and past consultation contents in the database.

[2282] "Suggestions" refers to the advice or solutions the system provides to the user.

[2283] "User Device" refers to the electronic device (e.g., smartphone, tablet, or PC) used by a User to access the System.

[2284] "Feedback" refers to the evaluations and opinions that users give regarding the suggestions provided.

[2285] A "machine learning model" refers to an algorithm that automatically learns from accumulated data and makes predictions and classifications.

[2286] "More precise support" refers to assistance that responds appropriately and effectively to the user's individual situation.

[2287] An "emotion score" is an index that numerically evaluates a user's emotional state based on their text input.

[2288] "Dynamically optimized food and drink" refers to food and drink suggestions that are tailored based on the user's emotions and physical condition.

[2289] "Generative AI model" refers to an artificial intelligence model used to perform a generative task.

[2290] "Emergency Support" means an emergency response required immediately by a User.

[2291] An "alert" is a notification that notifies you of an emergency or important information.

[2292] This invention is a system that allows users to anonymously and safely seek advice about their mental health and receive appropriate support. The system incorporates an emotion engine that recognizes the user's emotions and provides dynamically adjusted suggestions.

[2293] System configuration

[2294] This system includes a user device, a server, a database, a natural language processing engine, an emotion engine, and a machine learning model. The user device refers to an electronic device such as a smartphone, tablet, or PC, and the server is a computer system for receiving, processing, and transmitting data.

[2295] Program processing

[2296] Login and Access

[2297] A user accesses the system's website or application using their own device and enters their username and password to log in. This authentication information is sent to the server, and once the server has successfully authenticated, the user is provided with a chat interface for consultation.

[2298] Enter and submit your inquiry

[2299] The user enters the content of the consultation in text format into the chat interface, and this content is sent to the server.

[2300] Initial analysis and sentiment analysis

[2301] The server uses natural language processing technology to perform an initial analysis of the received consultation content, and then calculates an emotion score using an emotion engine, which allows the user's emotional state to be numerically evaluated.

[2302] Topic classification and historical data reference

[2303] Based on the analysis results, the consultation content is classified into topics, and similar consultation content and solutions are extracted by referring to a database of past consultation content.

[2304] Proposal generation and optimization

[2305] The server generates recommendations based on the extracted solutions and dynamically optimizes them based on the emotion score, suggesting foods and drinks that correspond to the user's emotions and physical condition.

[2306] View suggestions and get feedback

[2307] The generated suggestions are sent back to the user's device and displayed to the user. The user checks the suggestions and enters feedback. This feedback is sent to the server and stored in a database.

[2308] Training a machine learning model

[2309] The server uses the accumulated feedback to train a machine learning model and generate more accurate support for the next consultation.

[2310] Hardware and software used

[2311] Hardware: User devices (smartphones, tablets, PCs), servers (high-performance computing systems)

[2312] Software: Natural language processing engines (e.g., SpaCy, NLTK), emotion engines (e.g., Hugging Face's transformers library), machine learning models (e.g., Scikit-learn, TensorFlow)

[2313] Specific examples

[2314] Example 1:

[2315] If a user types, "I've been stressed out lately and have lost my appetite.":

[2316] The system suggests, "Would you like some relaxing herbal tea?"

[2317] The user orders the suggested herbal tea and enters feedback as "That was a good suggestion, I feel better."

[2318] Prompt Sentence Examples

[2319] "The user has recently been feeling stressed and has lost their appetite. Please suggest suitable foods and drinks. Use your historical database and emotion engine to generate the best suggestions."

[2320] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2321] Specific processing flow of the program

[2322] Step 1: User Login and Access

[2323] Input: The user enters their username and password from their device (smartphone, tablet, computer).

[2324] Data processing: The device sends the entered authentication information (username and password) to the server.

[2325] Data calculation: The server checks the authentication information against the database.

[2326] Output: If authentication is successful, the server starts a session and presents the user with a chat interface for consultation.

[2327] Step 2: Enter and submit your inquiry

[2328] Input: The user enters the content of their inquiry in text format into the chat interface.

[2329] Data processing: The device sends the consultation details entered to the server.

[2330] Data calculation: None (simple data transmission).

[2331] Output: The consultation content is received by the server.

[2332] Step 3: Initial analysis and sentiment analysis

[2333] Input: Text data of the consultation received by the server.

[2334] Data processing: The server performs initial analysis of the text data using a natural language processing engine (e.g., SpaCy, NLTK).

[2335] Data calculation: An emotion engine (e.g., Hugging Face transformers library) is used to calculate an emotion score and numerically evaluate the emotional state of the consultation.

[2336] Output: Analysis results of the text data and sentiment scores.

[2337] Step 4: Topic classification and historical data reference

[2338] Input: Initial analysis results and sentiment scores.

[2339] Data processing: The server categorizes the consultation content into topics based on the analysis results.

[2340] Data calculation: Query the database of past cons...

Claims

1. A means for users to input their consultation details in text format; a means for receiving the consultation content in a server and performing an initial analysis using natural language processing; A means for classifying the consultation contents into topics based on the initial analysis results; A means for searching a database of past consultation contents and extracting similar consultation contents and their solutions; a means for organizing the extracted solutions and generating proposals to be provided to a user; means for transmitting the generated proposal content to a user's terminal and displaying it to the user; means for receiving feedback from users and storing the feedback in a database; A means for training a machine learning model based on the accumulated data to generate more accurate support for the next consultation; A system including:

2. means for performing sentiment analysis in the initial analysis of the consultation content; means for calculating an emotion score in an initial analysis of the consultation content; The system of claim 1 further comprising:

3. 2. The system of claim 1, further comprising means for generating an alert if the generated suggestion is determined to require emergency support.

Images