System

Abstract

A system is provided.SOLUTION: A system, comprising: means for performing sentiment analysis based on user input sentiment and experience; means for generating empathetic feedback based on the sentiment analysis; and means for introducing to an expert as needed as a result of the feedback.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In modern society, people experiencing deep sadness or stress find it difficult to confide their feelings to those around them, resulting in feelings of loneliness and increased mental strain. It can be particularly difficult to receive appropriate support when facing significant stress, such as the death of a spouse, family member, or friend, or work pressure. This invention aims to provide such emotionally wounded people with a safe space where they can express their negative emotions at any time, helping them process their emotions and recover. [Means for solving the problem]

[0005] The present invention provides a system including means for performing sentiment analysis based on emotions and experiences input by a user, means for generating empathetic feedback based on the sentiment analysis, and means for referring the user to a specialist as needed as a result of the feedback. The sentiment analysis is performed using an algorithm trained on psychological principles, and empathetic and appropriate feedback is generated according to the analysis results. Furthermore, if the emotions are very severe, the system provides links or contact information for reliable professional counselors or support services, allowing the user to receive appropriate professional help.

[0006] A "user" is an individual who uses the system to input their feelings and experiences and receive feedback based on them.

[0007] "Sentiment analysis" is the process of analyzing the emotions and experiences input by a user to determine their emotional state.

[0008] "Empathetic feedback" refers to feedback that shows understanding of the user's feelings and provides comfort and encouragement using appropriate words.

[0009] "Expert" refers to an individual or organization with specialized knowledge and experience in the field of psychology or counseling.

[0010] "Referrals" refers to connecting users with trusted professionals and services when they need further specialized assistance.

[0011] "Principles of psychology" refers to the basic concepts and methodologies based on academic theories and empirical research in psychology.

[0012] An "algorithm" refers to a rule or method that defines the procedure for calculation or processing to achieve a specific purpose.

[0013] "System" refers to a set of components that process user-entered data and provide functions such as sentiment analysis, feedback generation, and referral to experts. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

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

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

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

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0029] 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.

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

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] Basic System Configuration

[0036] The system of the present invention consists of the following major components:

[0037] 1. Server

[0038] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0039] It takes input from the user and performs sentiment analysis.

[0040] Generate empathetic feedback based on analytics.

[0041] Generate specialist referrals as needed.

[0042] 2. Terminal

[0043] It provides an interface for users to input their emotions and experiences.

[0044] Send input data to the server.

[0045] View feedback and expert referrals from the server.

[0046] 3. Users

[0047] Input your emotions and experiences into the device.

[0048] Check for feedback and expert referrals from the server.

[0049] What the program does

[0050] server

[0051] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[0052] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[0053] 3. Once user input is received, it is fed into a sentiment analysis model to classify the sentiment as positive, negative, neutral, etc.

[0054] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[0055] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[0056] Terminal

[0057] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[0058] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[0059] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[0060] user

[0061] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[0062] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[0063] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[0064] Specific examples

[0065] A specific example of the system is shown below.

[0066] Example user input

[0067] "I've been very sad recently. A friend of mine passed away."

[0068] Server response example

[0069] Sentiment analysis result: Negative

[0070] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0071] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0072] Example of terminal display

[0073] "I feel your emotions. I understand your grief. I want to help you."

[0074] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0075] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[0076] The processing flow will be explained below.

[0077] Step 1:

[0078] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model for natural language processing (e.g., BERT), which prepares it for sentiment analysis.

[0079] Step 2:

[0080] The device provides a user interface that allows users to input their emotions and experiences. This interface consists of a simple text input field.

[0081] Step 3:

[0082] Users input their feelings and experiences into the device. For example, a user might input, "I've been feeling very sad recently. A friend of mine passed away."

[0083] Step 4:

[0084] The terminal sends the input data from the user to the server, which then receives the data in the appropriate structured format.

[0085] Step 5:

[0086] The server inputs the received user input text into a sentiment analysis model to analyze the sentiment, which classifies the input text as positive, negative, or neutral.

[0087] Step 6:

[0088] The server generates empathetic feedback based on the analysis results. For example, if the result of the sentiment analysis is negative, it generates feedback such as, "I feel your emotion. I understand your sadness."

[0089] Step 7:

[0090] The server generates referral information based on the intensity of the emotion. If the intensity of negative emotion exceeds a certain threshold, it generates a message providing contact information for an appropriate counselor or support service.

[0091] Step 8:

[0092] The server transmits the generated feedback and expert referral information to the terminal.

[0093] Step 9:

[0094] The device displays the feedback and referral information received from the server to the user, allowing them to feel understood and identify next steps to get the professional help they need.

[0095] Step 10:

[0096] The user can then review the feedback and referral information displayed on the device and, if necessary, contact a professional counselor or support service through the provided contact details, allowing the user to take concrete action to receive appropriate assistance.

[0097] Through the above processing steps, the system of the present invention properly analyzes the user's emotions and provides empathetic feedback and referrals to necessary specialists, thereby realizing psychological care.

[0098] Example 1

[0099] 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."

[0100] Traditional sentiment analysis systems lacked the ability to properly understand users' emotions and provide empathetic feedback. Furthermore, they often lacked the ability to refer users to appropriate experts when they were in a serious emotional situation. This often resulted in insufficient emotional care for users and a dissatisfying experience.

[0101] 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.

[0102] In this invention, the server includes means for performing sentiment analysis based on emotions and experiences input by a user, means for loading a sentiment analysis model using natural language processing and setting up a sentiment analysis pipeline, means for generating empathetic feedback based on the sentiment analysis, and means for generating referral information to an expert according to the intensity of the emotion. This makes it possible to analyze a user's emotions with high accuracy, provide empathetic feedback, and, if necessary, refer the user to an expert.

[0103] "User input" refers to the user sending their emotions and experiences to the system in text form via their terminal.

[0104] "Sentiment analysis" is the process of analyzing input text data and classifying the type of emotion (positive, negative, neutral, etc.) and intensity.

[0105] "Natural language processing" is a technology that enables computers to understand, generate, and analyze human language.

[0106] A "sentiment analysis model" is an AI model trained to perform sentiment analysis, such as BERT.

[0107] A "sentiment analysis pipeline" refers to a set of processes and algorithms for analyzing text data.

[0108] "Empathetic feedback" means providing messages that empathize with and show understanding of the user's feelings.

[0109] "Information on referrals to specialists" is information that provides consultation information and contact details for appropriate specialists or counselors depending on the intensity of the user's emotions.

[0110] "Terminal" means a device that allows a user to provide input and receive feedback.

[0111] The "server" is a central computing unit that hosts the sentiment analysis model, analyzes input data from users, and generates feedback.

[0112] A "tokenizer" is a part of natural language processing that breaks down text data into a more easily processable form.

[0113] MODE FOR CARRYING OUT THE INVENTION

[0114] Basic System Configuration

[0115] The system of the present invention consists of the following main components:

[0116] 1. Server

[0117] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0118] It takes input from the user and performs sentiment analysis.

[0119] Generate empathetic feedback based on analytics.

[0120] Generate specialist referrals as needed.

[0121] 2. Terminal

[0122] It provides an interface for users to input their emotions and experiences.

[0123] Send input data to the server.

[0124] View feedback and expert referrals from the server.

[0125] 3. Users

[0126] Input your emotions and experiences into the device.

[0127] Check for feedback and expert referrals from the server.

[0128] Specific server operations

[0129] During the initialization process, the server loads AI models for natural language processing (e.g., BERT) and tokenizers, which prepare the server to convert text data into an easily parseable format. Next, it sets up a sentiment analysis pipeline, preparing to analyze text data received from users with high accuracy.

[0130] When the server receives input from the user, it passes the content to a tokenizer, and the tokenized data is input into a sentiment analysis model. The sentiment analysis model classifies the sentiment of the text and generates an empathetic feedback message based on the results. For example, it generates a message such as, "I feel your emotion. I understand your sadness." It also generates referral information to a specialist if necessary, depending on the strength of the emotion. For example, it provides information such as, "I recommend that you consult a professional counselor."

[0131] Specific operation of the device

[0132] The device provides an interface that allows users to input their emotions and experiences. This interface includes text boxes and a submit button, allowing users to easily communicate their emotions to the system. When the user completes the input and clicks the submit button, the device sends the data to the server in an appropriate format.

[0133] The device receives feedback and referral information from the server and displays it to the user, allowing the user to feel that their feelings are understood and to obtain information on how to contact a specialist if necessary.

[0134] Specific user actions

[0135] Users input their feelings and experiences in text form through the device interface, for example, "I've been feeling very sad recently. A friend of mine passed away."

[0136] Once the user has completed and submitted the input, they will see the feedback generated by the server. This feedback is empathetic and appropriate, providing emotional support to the user. In the case of a serious emotional state, they will see information on referrals to specialists and, if necessary, consult with a specialist. This allows the user to receive further professional support.

[0137] Specific examples

[0138] A specific example of the system is shown below.

[0139] Example user input

[0140] "I've been very sad recently. A friend of mine passed away."

[0141] Server response example

[0142] Sentiment analysis result: Negative

[0143] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0144] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0145] Example of terminal display

[0146] "I feel your emotions. I understand your grief. I want to help you."

[0147] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0148] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[0149] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0150] Program processing flow

[0151] Step 1: Initialize the server

[0152] How it works:

[0153] The server performs the following initial settings when the system starts up.

[0154] Input: config file and model file path.

[0155] Data processing / data computation: The server loads the AI ​​model (e.g., BERT) and tokenizer from the specified directory and sets up the sentiment analysis pipeline.

[0156] Output: The initialized sentiment analysis pipeline.

[0157] Specific behavior: Loads the model file and tokenization settings, and initializes the sentiment analysis pipeline.

[0158] Step 2: Accepting input on the terminal

[0159] How it works:

[0160] Users input their emotions and experiences through the device's interface.

[0161] Input: Text data entered by the user.

[0162] Data processing / data calculation: The terminal receives the input and retrieves the string entered in the displayed text box.

[0163] Output: User-entered text data.

[0164] Specific behavior: The device displays a text box and a submit button, and the user clicks the "Submit" button after completing the input.

[0165] Step 3: Receiving and processing text on the server

[0166] How it works:

[0167] The server receives the text data sent from the terminal and prepares it for analysis.

[0168] Input: Text data sent from the terminal.

[0169] Data processing / data operation: The server passes the received text to the tokenizer and performs tokenization.

[0170] Output: Tokenized text data.

[0171] Specific operation: The server inputs text data into the tokenizer and obtains processed tokenized data.

[0172] Step 4: Sentiment analysis on the server

[0173] How it works:

[0174] The tokenized text data is fed into a sentiment analysis model to classify sentiment.

[0175] Input: Tokenized text data.

[0176] Data processing / data calculation: Using a sentiment analysis model, perform sentiment classification (positive, negative, neutral, etc.) of text data.

[0177] Output: Sentiment analysis results.

[0178] Specific operation: The server inputs the tokenized data into the sentiment analysis model and obtains the sentiment classification result.

[0179] Step 5: Generating feedback on the server

[0180] How it works:

[0181] Generate feedback messages for users based on the results of sentiment analysis.

[0182] Input: Sentiment analysis results.

[0183] Data processing / data calculation: Based on the results of sentiment analysis, generate empathetic feedback messages and referrals to experts, if necessary.

[0184] Output: Feedback message and expert referral information.

[0185] Specific operation: The server generates a feedback message based on the emotion classification result, which contains the necessary information for the user.

[0186] Step 6: Display feedback on your device

[0187] How it works:

[0188] The feedback message and expert introduction information are received from the server and displayed to the user.

[0189] Input: Feedback message and expert referral information from the server.

[0190] Data Processing / Data Calculation: The terminal displays the entered feedback message and referral information in an appropriate format.

[0191] Output: Display of feedback message and expert referral information.

[0192] Specific operation: The device displays the data received from the server on the user interface.

[0193] Step 7: Review user feedback and act

[0194] How it works:

[0195] Users can view the feedback displayed on their device and take action if necessary.

[0196] Input: Feedback message and expert referral information.

[0197] Data processing / data calculation: The user reads the feedback and considers any necessary actions.

[0198] Output: Action decision.

[0199] Specific Action: The user reads the displayed feedback message and contacts an expert if necessary.

[0200] Through this series of processes, the system is able to analyze the user's emotions with high accuracy and provide appropriate feedback and support information.

[0201] (Application example 1)

[0202] 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."

[0203] Conventional emotion analysis systems only analyze a user's emotional state, but do not necessarily provide sufficient feedback for subsequent responses. Furthermore, they lack a mechanism for quickly taking appropriate measures when a user is experiencing stress or anxiety. In particular, when a user is experiencing high levels of stress or anxiety, it is necessary to recognize the situation early and provide appropriate support.

[0204] 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.

[0205] In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for providing an emotion input interface, and means for issuing a warning if the user's emotional state is high in stress or anxiety based on the analysis results. This makes it possible to analyze the emotions and experiences input by the user with high accuracy and provide appropriate feedback and take early measures.

[0206] "User" refers to a person who uses the system.

[0207] An "emotion input interface" is an interface that allows users to input their own emotions and experiences in text format.

[0208] "Sentiment analysis" is the process of analyzing text entered by a user and identifying the type of sentiment (positive, negative, neutral) from its content.

[0209] "Empathetic feedback" refers to messages that understand the user's emotions based on the results of sentiment analysis and respond in a way that is sympathetic to those emotions.

[0210] "Professional Referrals" refers to information that connects users to appropriate professional counselors and support services when a user is deemed to be experiencing a serious emotional state.

[0211] The "means for issuing an alert" is a means for notifying the user or other relevant parties when the user's emotional state is one of high stress or anxiety.

[0212] "Server" refers to the computer system that performs core processing such as sentiment analysis and feedback generation.

[0213] The system configuration for implementing this invention is mainly divided into two main components: a server and a terminal.

[0214] Server configuration and functions

[0215] The server analyzes the text data entered by the user based on their emotions and experiences, and generates empathetic feedback based on the results. It also provides referral information to specialists as needed. Specifically, the system uses the following hardware and software:

[0216] Hardware

[0217] Computer system: Performs core processing for sentiment analysis and feedback generation.

[0218] software

[0219] BertTokenizer: Tokenizes user-entered text and converts it into a format suitable for analysis models.

[0220] BertForSequenceClassification: Performs sentiment analysis based on tokenized input and outputs sentiment classification results.

[0221] Softmax function: Calculates the probability of each emotion based on the classification results.

[0222] Device configuration and functions

[0223] The terminal provides an interface for the user to input emotions and experiences, and receives and displays feedback from the server and information on referrals to experts.

[0224] Hardware

[0225] Smartphone: A device for providing a user interface.

[0226] software

[0227] Dedicated application: An application for emotion input interface and communication with the server.

[0228] User Actions

[0229] Users access a dedicated application using their smartphone and enter their current feelings and experiences in text format, for example, "I've been feeling very sad recently. A friend of mine passed away."

[0230] The input data is sent to the server, which tokenizes the text using BertTokenizer and performs sentiment analysis using BertForSequenceClassification. Based on the analysis results, the probability of each emotion is calculated using a softmax function, and empathetic feedback is generated based on the results. For example, a message such as "I feel your emotions. I understand your sadness. I want to help you" is generated.

[0231] Based on the intensity of the emotion, referral information is also generated if necessary, and this information is presented to the user in the form of, "Serious emotions have been detected. We recommend that you consult a professional counselor."

[0232] Through the above process, the present invention can understand the user's emotions with high accuracy and provide appropriate feedback and early countermeasures.

[0233] Prompt Sentence Examples

[0234] "I've been feeling very sad lately. A friend passed away. Please analyze my current emotions and generate appropriate feedback."

[0235] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0236] Step 1:

[0237] The user uses a device to input their emotions and experiences in text form into the emotion input interface. For example, they might input, "I've been feeling very sad recently. My friend passed away." This input becomes input data for subsequent processing.

[0238] Step 2:

[0239] The terminal transmits the user's input data to the server, which includes specific operations such as packaging the data in an appropriate format and sending it to the server, for example, converting it to JSON format and sending it.

[0240] Step 3:

[0241] The server receives the user's input data and tokenizes it using BertTokenizer. Specifically, it splits the input text data into words and formats them. The output of this step is the tokenized input data.

[0242] Step 4:

[0243] The server performs sentiment analysis by inputting the tokenized input data into the BertForSequenceClassification model. This model analyzes the tokenized data and outputs the probability of each emotion (positive, negative, neutral, etc.). The output of this step is a list containing the probability of each emotion.

[0244] Step 5:

[0245] The server normalizes the resulting probability list using a softmax function and identifies the emotion with the highest probability as the emotion classification result. Specifically, given the probabilities for multiple emotions, the server selects the one with the maximum value. The output of this step is the emotion classification result (e.g., negative).

[0246] Step 6:

[0247] The server generates an empathetic feedback message based on the emotion classification result. For example, if a negative emotion is detected, it generates a message such as "I feel your emotion. I understand your sadness." The output of this step is the feedback message.

[0248] Step 7:

[0249] The server generates a referral to a specialist based on the intensity of the emotion. If a highly negative emotion is detected, it generates a referral such as "Serious emotions have been detected. We recommend that you consult a professional counselor." The output of this step is a referral to a specialist.

[0250] Step 8:

[0251] The server transmits the generated feedback message and expert referral information to the terminal, which includes specific operations of converting the data into a suitable format and transmitting it over the network.

[0252] Step 9:

[0253] The device displays the received feedback message and referral information to the expert to the user. For example, it displays a message on the screen saying, "I feel your emotions. I understand your sadness. I want to help you."

[0254] Step 10:

[0255] The user checks the displayed feedback message and information on referrals to experts, and consults with an expert if necessary.

[0256] 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.

[0257] Basic System Configuration

[0258] The system of the present invention consists of the following major components:

[0259] 1. Server

[0260] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0261] It takes input from the user and performs sentiment analysis.

[0262] Generate empathetic feedback based on analytics.

[0263] Generate specialist referrals as needed.

[0264] Use an emotion engine to identify emotions in real time in response to user input.

[0265] 2. Terminal

[0266] It provides an interface for users to input their emotions and experiences.

[0267] Send input data to the server.

[0268] View feedback and expert referrals from the server.

[0269] 3. Users

[0270] Input your emotions and experiences into the device.

[0271] Check for feedback and expert referrals from the server.

[0272] What the program does

[0273] server

[0274] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[0275] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[0276] 3. Once the user input is received, it is fed into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the input text as positive, negative, or neutral.

[0277] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[0278] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[0279] 6. Use an emotion engine to identify real-time emotions in response to user input. This emotion engine uses machine learning models to classify emotions with high accuracy, and in some cases also combines speech recognition technology to identify emotions.

[0280] Terminal

[0281] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[0282] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[0283] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[0284] user

[0285] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[0286] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[0287] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[0288] Specific examples

[0289] A specific example of the system is shown below.

[0290] Example user input

[0291] "I've been very sad recently. A friend of mine passed away."

[0292] Server response example

[0293] Sentiment analysis result: Negative

[0294] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0295] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0296] Example of terminal display

[0297] "I feel your emotions. I understand your grief. I want to help you."

[0298] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0299] This system provides users with a safe environment where they can express their emotions. They can receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide them to seek professional help.

[0300] The processing flow will be explained below.

[0301] Step 1:

[0302] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model used for natural language processing (e.g., BERT).

[0303] Step 2:

[0304] The server sets up an emotion engine, which uses machine learning models to recognize user emotions in real time.

[0305] Step 3:

[0306] The device displays an interface for users to input their emotions and experiences, providing UI elements such as input fields and buttons to allow users to easily describe their emotions.

[0307] Step 4:

[0308] The user inputs their feelings and experiences into the device. For example, the user inputs text such as, "I've been feeling very sad recently. A friend of mine passed away."

[0309] Step 5:

[0310] The terminal sends user input in real time to the server, which then passes this data to the server in the appropriate format.

[0311] Step 6:

[0312] The server inputs the received text into a sentiment analysis model to perform an initial sentiment analysis, which classifies the user's input text as positive, negative, or neutral.

[0313] Step 7:

[0314] The server uses an emotion engine to monitor changes in emotions in real time. As the user continues to input, the engine analyzes the emotions and prepares the most appropriate feedback.

[0315] Step 8:

[0316] The server generates empathetic feedback based on the final sentiment analysis result. For example, if the sentiment analysis result is negative, the server prepares feedback such as "I feel your emotion. I understand your sadness."

[0317] Step 9:

[0318] The server generates referral information for specialists based on the strength of the user's emotions. If the emotions are strong, it provides contact information for appropriate counselors and support services.

[0319] Step 10:

[0320] The server transmits the generated feedback and referral information to the expert to the terminal.

[0321] Step 11:

[0322] The terminal displays the feedback and introduction information received from the server to the user, allowing the user to receive appropriate feedback in real time.

[0323] Step 12:

[0324] Users can view the feedback displayed on their device and, if needed, seek further assistance using the provided contact details of a specialist, gaining reassurance that their feelings are understood and the immediate support they need.

[0325] Through this process, the system of the present invention analyzes the user's emotions in real time, provides empathetic feedback, and quickly refers the user to a specialist if necessary.

[0326] Example 2

[0327] 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."

[0328] Conventional emotion analysis systems have had difficulty accurately analyzing users' emotions and providing empathetic feedback based on the results. Furthermore, there was a lack of systems that could identify emotions in real time and provide referral information to specialists as needed, making it difficult to provide prompt and appropriate care for users' mental health. To solve this issue, a system capable of highly accurate emotion analysis and providing empathetic feedback in real time was needed.

[0329] 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.

[0330] In this invention, the server includes means for loading an artificial intelligence model and a tokenizer in an initialization process, means for setting up a sentiment analysis pipeline and preparing to analyze text data, means for receiving input from a user and inputting the content of the input into the sentiment analysis model, means for classifying the received input text as positive, negative, or neutral, means for generating empathetic feedback based on the analysis results, means for generating referral information to an expert if necessary depending on the strength of the emotion, and means for using an emotion engine that identifies emotions in real time, thereby enabling highly accurate analysis of user emotions, providing empathetic feedback in real time, and providing referral information to an appropriate expert if necessary.

[0331] The "initialization process" is a series of steps that loads the artificial intelligence model and tokenizer and prepares the system.

[0332] An "artificial intelligence model" is a set of pre-trained mathematical algorithms for natural language processing, including, for example, BERT.

[0333] A "tokenizer" is a tool or algorithm that breaks text into pieces that are easier to parse.

[0334] A "sentiment analysis pipeline" is the series of processing steps required to take a user's input text and analyze its sentiment.

[0335] "Text data" refers to data in the form of text that a user inputs into the system.

[0336] A "sentiment analysis model" is an algorithm that analyzes text data and classifies it as positive, negative, or neutral.

[0337] "Empathetic feedback" is a response message that is appropriate and empathetic to the user's emotional state.

[0338] "Professional Referral Information" is contact information for counselors or support services provided to users for further professional help.

[0339] "Real-time emotion identification" is the process of instantly analyzing and identifying emotions in response to user input.

[0340] An "emotion engine" is a collection of machine learning models and techniques for analyzing and identifying emotions in real time with high accuracy.

[0341] MODE FOR CARRYING OUT THE INVENTION

[0342] The system for implementing the present invention mainly comprises three components: a server, a terminal, and a user.

[0343] server

[0344] The server is responsible for central control of the system. First, the server loads the AI ​​model and tokenizer during the system initialization process. This AI model uses a pre-trained model for natural language processing, such as BERT. The tokenizer is a tool for dividing text data entered by the user into units that are easy to analyze.

[0345] Once loaded, the server sets up a sentiment analysis pipeline, ready to parse the text data from the user, using a tokenizer to convert the text into a form suitable for the model.

[0346] Upon receiving input data from the user, the server inputs the data into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the text data as positive, negative, or neutral, identifying the user's emotional state. Based on the analysis results, the server then generates an empathetic feedback message. For example, a message such as "I feel your emotions. I understand your sadness."

[0347] Depending on the strength of the emotion, it also generates referral information to a specialist. For example, if a severe negative emotion is detected, it will provide a referral such as "We recommend you consult a professional counselor." In addition, it uses an emotion engine that identifies emotions in real time, instantly analyzing and identifying the emotion in response to user input.

[0348] Terminal

[0349] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, through which users can communicate their emotions to the system. The input data is converted into an appropriate format (e.g., JSON format) and sent to the server.

[0350] The feedback received from the server and information on referrals to experts are clearly displayed on the device interface, with appropriate font sizes and colors to ensure that users can immediately understand the feedback.

[0351] user

[0352] Users input their emotions and experiences through the device interface. Specifically, they can enter detailed emotions and background information in text or voice format. After completing the input, they check the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional care for the user.

[0353] Furthermore, if the emotional state is serious, the user can check for referral information and consult with a specialist if necessary, allowing the user to receive further professional support.

[0354] Specific examples

[0355] A specific example of the system is shown below.

[0356] Example user input

[0357] "I've been very sad recently. A friend of mine passed away."

[0358] Server response example

[0359] Sentiment analysis result: Negative

[0360] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0361] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0362] Example of terminal display

[0363] "I feel your emotions. I understand your grief. I want to help you."

[0364] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0365] This system allows users to safely express their emotions and receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide users to seek professional help.

[0366] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0367] Server Processing

[0368] Step 1: Initialization process

[0369] The server loads the artificial intelligence model and tokenizer when the system starts up. Specifically, it loads a pre-trained natural language processing model such as BERT into memory and initializes the tokenizer. The input to this step is the required model and tokenizer files, and the output is the loaded model and tokenizer ready for use.

[0370] Step 2: Setting up the sentiment analysis pipeline

[0371] The server sets up a sentiment analysis pipeline, which uses a tokenizer to convert text data into tokens and format them for input to the model. For example, the tokenizer splits the text into words and subwords, which are then converted into tensors for input to the model. The input is raw text data, and the output is tokenized data in a format that the model can parse.

[0372] Step 3: Receiving User Input

[0373] The server receives the user's text data sent from the device. Once received, it parses the input data in a format such as JSON and stores it in a data structure. The input of this step is the text data entered by the user, and the output is a data structure that is ready for analysis.

[0374] Step 4: Perform sentiment analysis

[0375] The server inputs the received text data into a sentiment analysis model to classify the sentiment. Specifically, the model classifies the text as positive, negative, or neutral and calculates the probability of each category. The input is tokenized text data, and the output is the sentiment classification result (e.g., 50% positive, 40% negative, 10% neutral).

[0376] Step 5: Generate feedback

[0377] The server generates empathetic feedback based on the emotion analysis results. For example, if the emotion is strong and negative, it generates a message such as "I feel your emotion. I understand your sadness." The input is the emotion analysis results, and the output is the feedback message displayed to the user.

[0378] Step 6: Generate expert referrals

[0379] Depending on the strength of the emotion, the server generates referral information to a specialist. For example, if a severe negative emotion is detected, the server will provide a message saying, "We recommend that you consult a professional counselor," along with contact information. The input is the emotion analysis result, and the output is a message including the contact information of the specialist.

[0380] Step 7: Real-time emotion identification

[0381] The server uses an emotion engine to identify emotions in real time. When there is voice input, it uses speech recognition technology to convert the voice into text and perform emotion classification. The input is real-time voice or text data, and the output is the immediate emotion classification result.

[0382] Terminal handling

[0383] Step 1: Provide an input interface

[0384] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, which users use to input data. The input is information about the user's emotions and experiences, and the output is the input data in an organized format.

[0385] Step 2: Submitting input data

[0386] The terminal converts the user input into an appropriate format and sends it to the server. For example, it encodes text data into JSON format and sends it to the server via an HTTP request. The input is the user's input data, and the output is the data sent to the server.

[0387] Step 3: View your feedback

[0388] The feedback received from the server and the referral information to the expert are displayed to the user. Specifically, the feedback messages and referral information are displayed in an easy-to-understand interface. The input is the received feedback and referral information, and the output is the message displayed to the user.

[0389] User Action

[0390] Step 1: Input your emotions and experiences

[0391] Users input their emotions and experiences through the device interface. Specifically, they input emotions and background information in text or voice format. The input is information about the user's emotions and experiences, and the output is the data entered into the device.

[0392] Step 2: Review feedback

[0393] After completing the input, the user confirms the feedback generated by the server. The feedback is composed of empathetic and appropriate content, allowing the user to confirm it and feel that their feelings have been understood. The input is the feedback from the server, and the output is the confirmed feedback content.

[0394] Step 3: Confirm specialist referral information

[0395] In the case of a severe emotional state, referral information to a specialist is confirmed and, if necessary, a specialist is consulted, allowing the user to receive further professional assistance. The input is the referral information from the server, and the output is the confirmed specialist's contact information.

[0396] (Application example 2)

[0397] 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."

[0398] The present invention aims to provide a system that can provide both mental care and entertainment to users by not only providing appropriate feedback based on the emotions and experiences input by the user but also recommending appropriate content according to the user's emotional state.

[0399] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for recommending appropriate content based on the user's emotional state, and means for providing the content to the user. This allows the user to not only receive feedback according to their own emotional state, but also to view appropriate content.

[0400] "Sentiment analysis" is a technology that detects emotions from text data entered by the user and classifies the type of emotion (positive, negative, neutral).

[0401] "Feedback" is an empathetic message that is returned to the user based on the results of sentiment analysis, intended to understand and support the user's emotional state.

[0402] "Professional Referrals" is a way to provide links and contact information for trusted professional counselors and support services when sentiment analysis indicates a user is experiencing a serious emotional state.

[0403] "Content recommendation" is a technology that selects and provides appropriate entertainment and information such as videos and text based on the user's emotional state.

[0404] "Server" is a computer system that performs the functions of sentiment analysis, feedback generation, expert referrals, and content recommendation, and manages the overall system.

[0405] "Entertainment" refers to media such as videos, music, and games that provide users with fun and relaxation.

[0406] "User interface" refers to the means of interaction that allows users to input their emotions and experiences into the system, and to view feedback and recommended content.

[0407] "Principles of psychology" is a general term for theories and laws that provide the academic foundation for more reliable understanding and classification of emotions.

[0408] An "algorithm" is a set of procedures or computational methods for performing sentiment analysis, feedback generation, or content recommendation.

[0409] Basic System Configuration

[0410] The system of the present invention consists of the following major components:

[0411] 1. Server

[0412] The server loads AI models and tokenizers and sets up a sentiment analysis pipeline, specifically using popular models for natural language processing (e.g., BERT).

[0413] It takes input from the user and performs sentiment analysis: the server classifies the input text as positive, negative, or neutral.

[0414] Generate empathetic feedback based on the analysis, such as "I feel your emotions. I understand your sadness."

[0415] Depending on the intensity of the emotion, the system generates referral information to a specialist if necessary. For example, if a severe negative emotion is detected, the system will provide a referral such as "We recommend that you consult a professional counselor."

[0416] It recommends appropriate content based on the user's emotional state, including relaxation videos, healing music, documentaries, and more.

[0417] 2. Terminal

[0418] The device provides an interface for users to input their emotions and experiences, and through this interface users can communicate their emotions to the system.

[0419] User input is sent to the server, where it is properly formatted for sentiment analysis and feedback generation.

[0420] The system receives feedback from the server, information on referrals to experts, and recommended content, and displays it to the user. The user feels that their feelings are understood and can consult with an expert if necessary.

[0421] What the program does

[0422] server

[0423] During the initialization process, the server loads AI models and tokenizers, including natural language processing models (e.g., BERT and RoBERTa). It then sets up a sentiment analysis pipeline and prepares to analyze incoming text data. When it receives user input, it feeds the content into the sentiment analysis model to classify the sentiment. Based on the results of the sentiment analysis, it generates empathetic feedback and, if necessary, a referral to an expert. It also recommends appropriate content based on the user's emotional state.

[0424] Terminal

[0425] The device provides an interface where users can input their emotions and experiences, and sends the input to the server. The server then receives feedback, referral information to experts, and recommended content, which are then displayed to the user. This allows users to feel that their emotions are understood and to consult with experts if necessary.

[0426] user

[0427] Users enter their emotions and experiences in text format through the device interface. After completing the input, they check the feedback and recommended content generated by the server. In the case of a serious emotional state, they check the information for referrals to specialists and consult with them if necessary.

[0428] Specific examples

[0429] If a user types "I feel very tired today," the server will classify this input as "NEGATIVE" and recommend content such as healing music or meditation guides. Here are some examples of specific prompts:

[0430] python

[0431] from transformers import pipeline

[0432] emotion_pipeline = pipeline("sentiment-analysis")

[0433] User inputs emotion

[0434] user_input = "I'm very tired today"

[0435] Perform sentiment analysis

[0436] emotion = emotion_pipeline(user_input)

[0437] print(emotion)

[0438] By using this prompt, you can see the process of obtaining the sentiment analysis results.

[0439] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0440] Step 1:

[0441] The device provides an interface that allows users to input their emotions and experiences. When the user inputs something, the text data is sent to the device. The input data is text data that appropriately expresses the user's emotional state. This data is important for accurately reflecting the user's current situation.

[0442] Step 2:

[0443] The device sends the text data received from the user to the server in the appropriate format, taking care not to corrupt the data. The format is a standard data format such as JSON. This data transmission is a prerequisite for the server to accurately perform subsequent sentiment analysis.

[0444] Step 3:

[0445] The server tokenizes the received text data to input into the AI ​​model. Tokenization splits the text into words or tokens and converts it into a format that the AI ​​model can understand. This process makes it easier for the model to understand each part of the text.

[0446] Step 4:

[0447] The server inputs the tokenized data into a sentiment analysis model to classify the sentiment. This uses a generative AI model (e.g., BERT or RoBERTa) to classify sentiment as positive, negative, or neutral. This model performs highly accurate sentiment analysis based on the input data.

[0448] Step 5:

[0449] The server generates an empathetic feedback message based on the emotion classification results. For example, if a negative emotion is detected, the server generates a message saying, "I feel your emotion. I understand your sadness." This feedback helps users feel understood.

[0450] Step 6:

[0451] The server generates referral information to a specialist as needed depending on the intensity of the emotion. If a severe negative emotion is detected, the server will provide information such as "We recommend you consult a professional counselor." This allows the user to receive the necessary professional help.

[0452] Step 7:

[0453] The server then recommends content appropriate to the user's emotional state based on the results of the emotion analysis. For example, if a user inputs that they are tired, it will recommend relaxation videos or soothing music. This process is intended to provide appropriate entertainment according to the user's emotional state.

[0454] Step 8:

[0455] The server then sends the generated feedback messages, expert referrals, and recommended content to the device, formatting the data to be displayed appropriately to the user. The data is then displayed accurately on the user's device.

[0456] Step 9:

[0457] The device displays the feedback, referral information, and recommended content received from the server to the user. The user can confirm this information and feel that their feelings are understood. They can also access experts or watch recommended content as needed. This step ensures that the user receives appropriate support and entertainment from the system.

[0458] 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.

[0459] 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.

[0460] 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.

[0461] [Second embodiment]

[0462] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0463] 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.

[0464] 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).

[0465] 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.

[0466] 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.

[0467] 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).

[0468] 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.

[0469] 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.

[0470] 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.

[0471] 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.

[0472] 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.

[0473] 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."

[0474] Basic System Configuration

[0475] The system of the present invention consists of the following major components:

[0476] 1. Server

[0477] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0478] It takes input from the user and performs sentiment analysis.

[0479] Generate empathetic feedback based on analytics.

[0480] Generate specialist referrals as needed.

[0481] 2. Terminal

[0482] It provides an interface for users to input their emotions and experiences.

[0483] Send input data to the server.

[0484] View feedback and expert referrals from the server.

[0485] 3. Users

[0486] Input your emotions and experiences into the device.

[0487] Check for feedback and expert referrals from the server.

[0488] What the program does

[0489] server

[0490] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[0491] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[0492] 3. Once user input is received, it is fed into a sentiment analysis model to classify the sentiment as positive, negative, neutral, etc.

[0493] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[0494] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[0495] Terminal

[0496] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[0497] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[0498] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[0499] user

[0500] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[0501] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[0502] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[0503] Specific examples

[0504] A specific example of the system is shown below.

[0505] Example user input

[0506] "I've been very sad recently. A friend of mine passed away."

[0507] Server response example

[0508] Sentiment analysis result: Negative

[0509] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0510] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0511] Example of terminal display

[0512] "I feel your emotions. I understand your grief. I want to help you."

[0513] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0514] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[0515] The processing flow will be explained below.

[0516] Step 1:

[0517] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model for natural language processing (e.g., BERT), which prepares it for sentiment analysis.

[0518] Step 2:

[0519] The device provides a user interface that allows users to input their emotions and experiences. This interface consists of a simple text input field.

[0520] Step 3:

[0521] Users input their feelings and experiences into the device. For example, a user might input, "I've been feeling very sad recently. A friend of mine passed away."

[0522] Step 4:

[0523] The terminal sends the input data from the user to the server, which then receives the data in the appropriate structured format.

[0524] Step 5:

[0525] The server inputs the received user input text into a sentiment analysis model to analyze the sentiment, which classifies the input text as positive, negative, or neutral.

[0526] Step 6:

[0527] The server generates empathetic feedback based on the analysis results. For example, if the result of the sentiment analysis is negative, it generates feedback such as, "I feel your emotion. I understand your sadness."

[0528] Step 7:

[0529] The server generates referral information based on the intensity of the emotion. If the intensity of negative emotion exceeds a certain threshold, it generates a message providing contact information for an appropriate counselor or support service.

[0530] Step 8:

[0531] The server transmits the generated feedback and expert referral information to the terminal.

[0532] Step 9:

[0533] The device displays the feedback and referral information received from the server to the user, allowing them to feel understood and identify next steps to get the professional help they need.

[0534] Step 10:

[0535] The user can then review the feedback and referral information displayed on the device and, if necessary, contact a professional counselor or support service through the provided contact details, allowing the user to take concrete action to receive appropriate assistance.

[0536] Through the above processing steps, the system of the present invention properly analyzes the user's emotions and provides empathetic feedback and referrals to necessary specialists, thereby realizing psychological care.

[0537] Example 1

[0538] 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."

[0539] Traditional sentiment analysis systems lacked the ability to properly understand users' emotions and provide empathetic feedback. Furthermore, they often lacked the ability to refer users to appropriate experts when they were in a serious emotional situation. This often resulted in insufficient emotional care for users and a dissatisfying experience.

[0540] 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.

[0541] In this invention, the server includes means for performing sentiment analysis based on emotions and experiences input by a user, means for loading a sentiment analysis model using natural language processing and setting up a sentiment analysis pipeline, means for generating empathetic feedback based on the sentiment analysis, and means for generating referral information to an expert according to the intensity of the emotion. This makes it possible to analyze a user's emotions with high accuracy, provide empathetic feedback, and, if necessary, refer the user to an expert.

[0542] "User input" refers to the user sending their emotions and experiences to the system in text form via their terminal.

[0543] "Sentiment analysis" is the process of analyzing input text data and classifying the type of emotion (positive, negative, neutral, etc.) and intensity.

[0544] "Natural language processing" is a technology that enables computers to understand, generate, and analyze human language.

[0545] A "sentiment analysis model" is an AI model trained to perform sentiment analysis, such as BERT.

[0546] A "sentiment analysis pipeline" refers to a set of processes and algorithms for analyzing text data.

[0547] "Empathetic feedback" means providing messages that empathize with and show understanding of the user's feelings.

[0548] "Information on referrals to specialists" is information that provides consultation information and contact details for appropriate specialists or counselors depending on the intensity of the user's emotions.

[0549] "Terminal" means a device that allows a user to provide input and receive feedback.

[0550] The "server" is a central computing unit that hosts the sentiment analysis model, analyzes input data from users, and generates feedback.

[0551] A "tokenizer" is a part of natural language processing that breaks down text data into a more easily processable form.

[0552] MODE FOR CARRYING OUT THE INVENTION

[0553] Basic System Configuration

[0554] The system of the present invention consists of the following main components:

[0555] 1. Server

[0556] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0557] It takes input from the user and performs sentiment analysis.

[0558] Generate empathetic feedback based on analytics.

[0559] Generate specialist referrals as needed.

[0560] 2. Terminal

[0561] It provides an interface for users to input their emotions and experiences.

[0562] Send input data to the server.

[0563] View feedback and expert referrals from the server.

[0564] 3. Users

[0565] Input your emotions and experiences into the device.

[0566] Check for feedback and expert referrals from the server.

[0567] Specific server operations

[0568] During the initialization process, the server loads AI models for natural language processing (e.g., BERT) and tokenizers, which prepare the server to convert text data into an easily parseable format. Next, it sets up a sentiment analysis pipeline, preparing to analyze text data received from users with high accuracy.

[0569] When the server receives input from the user, it passes the content to a tokenizer, and the tokenized data is input into a sentiment analysis model. The sentiment analysis model classifies the sentiment of the text and generates an empathetic feedback message based on the results. For example, it generates a message such as, "I feel your emotion. I understand your sadness." It also generates referral information to a specialist if necessary, depending on the strength of the emotion. For example, it provides information such as, "I recommend that you consult a professional counselor."

[0570] Specific operation of the device

[0571] The device provides an interface that allows users to input their emotions and experiences. This interface includes text boxes and a submit button, allowing users to easily communicate their emotions to the system. When the user completes the input and clicks the submit button, the device sends the data to the server in an appropriate format.

[0572] The device receives feedback and referral information from the server and displays it to the user, allowing the user to feel that their feelings are understood and to obtain information on how to contact a specialist if necessary.

[0573] Specific user actions

[0574] Users input their feelings and experiences in text form through the device interface, for example, "I've been feeling very sad recently. A friend of mine passed away."

[0575] Once the user has completed and submitted the input, they will see the feedback generated by the server. This feedback is empathetic and appropriate, providing emotional support to the user. In the case of a serious emotional state, they will see information on referrals to specialists and, if necessary, consult with a specialist. This allows the user to receive further professional support.

[0576] Specific examples

[0577] A specific example of the system is shown below.

[0578] Example user input

[0579] "I've been very sad recently. A friend of mine passed away."

[0580] Server response example

[0581] Sentiment analysis result: Negative

[0582] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0583] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0584] Example of terminal display

[0585] "I feel your emotions. I understand your grief. I want to help you."

[0586] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0587] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[0588] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0589] Program processing flow

[0590] Step 1: Initialize the server

[0591] How it works:

[0592] The server performs the following initial settings when the system starts up.

[0593] Input: config file and model file path.

[0594] Data processing / data computation: The server loads the AI ​​model (e.g., BERT) and tokenizer from the specified directory and sets up the sentiment analysis pipeline.

[0595] Output: The initialized sentiment analysis pipeline.

[0596] Specific behavior: Loads the model file and tokenization settings, and initializes the sentiment analysis pipeline.

[0597] Step 2: Accepting input on the terminal

[0598] How it works:

[0599] Users input their emotions and experiences through the device's interface.

[0600] Input: Text data entered by the user.

[0601] Data processing / data calculation: The terminal receives the input and retrieves the string entered in the displayed text box.

[0602] Output: User-entered text data.

[0603] Specific behavior: The device displays a text box and a submit button, and the user clicks the "Submit" button after completing the input.

[0604] Step 3: Receiving and processing text on the server

[0605] How it works:

[0606] The server receives the text data sent from the terminal and prepares it for analysis.

[0607] Input: Text data sent from the terminal.

[0608] Data processing / data operation: The server passes the received text to the tokenizer and performs tokenization.

[0609] Output: Tokenized text data.

[0610] Specific operation: The server inputs text data into the tokenizer and obtains processed tokenized data.

[0611] Step 4: Sentiment analysis on the server

[0612] How it works:

[0613] The tokenized text data is fed into a sentiment analysis model to classify sentiment.

[0614] Input: Tokenized text data.

[0615] Data processing / data calculation: Using a sentiment analysis model, perform sentiment classification (positive, negative, neutral, etc.) of text data.

[0616] Output: Sentiment analysis results.

[0617] Specific operation: The server inputs the tokenized data into the sentiment analysis model and obtains the sentiment classification result.

[0618] Step 5: Generating feedback on the server

[0619] How it works:

[0620] Generate feedback messages for users based on the results of sentiment analysis.

[0621] Input: Sentiment analysis results.

[0622] Data processing / data calculation: Based on the results of sentiment analysis, generate empathetic feedback messages and referrals to experts, if necessary.

[0623] Output: Feedback message and expert referral information.

[0624] Specific operation: The server generates a feedback message based on the emotion classification result, which contains the necessary information for the user.

[0625] Step 6: Display feedback on your device

[0626] How it works:

[0627] The feedback message and expert introduction information are received from the server and displayed to the user.

[0628] Input: Feedback message and expert referral information from the server.

[0629] Data Processing / Data Calculation: The terminal displays the entered feedback message and referral information in an appropriate format.

[0630] Output: Display of feedback message and expert referral information.

[0631] Specific operation: The device displays the data received from the server on the user interface.

[0632] Step 7: Review user feedback and act

[0633] How it works:

[0634] Users can view the feedback displayed on their device and take action if necessary.

[0635] Input: Feedback message and expert referral information.

[0636] Data processing / data calculation: The user reads the feedback and considers any necessary actions.

[0637] Output: Action decision.

[0638] Specific Action: The user reads the displayed feedback message and contacts an expert if necessary.

[0639] Through this series of processes, the system is able to analyze the user's emotions with high accuracy and provide appropriate feedback and support information.

[0640] (Application example 1)

[0641] 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."

[0642] Conventional emotion analysis systems only analyze a user's emotional state, but do not necessarily provide sufficient feedback for subsequent responses. Furthermore, they lack a mechanism for quickly taking appropriate measures when a user is experiencing stress or anxiety. In particular, when a user is experiencing high levels of stress or anxiety, it is necessary to recognize the situation early and provide appropriate support.

[0643] 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.

[0644] In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for providing an emotion input interface, and means for issuing a warning if the user's emotional state is high in stress or anxiety based on the analysis results. This makes it possible to analyze the emotions and experiences input by the user with high accuracy and provide appropriate feedback and take early measures.

[0645] "User" refers to a person who uses the system.

[0646] An "emotion input interface" is an interface that allows users to input their own emotions and experiences in text format.

[0647] "Sentiment analysis" is the process of analyzing text entered by a user and identifying the type of sentiment (positive, negative, neutral) from its content.

[0648] "Empathetic feedback" refers to messages that understand the user's emotions based on the results of sentiment analysis and respond in a way that is sympathetic to those emotions.

[0649] "Professional Referrals" refers to information that connects users to appropriate professional counselors and support services when a user is deemed to be experiencing a serious emotional state.

[0650] The "means for issuing an alert" is a means for notifying the user or other relevant parties when the user's emotional state is one of high stress or anxiety.

[0651] "Server" refers to the computer system that performs core processing such as sentiment analysis and feedback generation.

[0652] The system configuration for implementing this invention is mainly divided into two main components: a server and a terminal.

[0653] Server configuration and functions

[0654] The server analyzes the text data entered by the user based on their emotions and experiences, and generates empathetic feedback based on the results. It also provides referral information to specialists as needed. Specifically, the system uses the following hardware and software:

[0655] Hardware

[0656] Computer system: Performs core processing for sentiment analysis and feedback generation.

[0657] software

[0658] BertTokenizer: Tokenizes user-entered text and converts it into a format suitable for analysis models.

[0659] BertForSequenceClassification: Performs sentiment analysis based on tokenized input and outputs sentiment classification results.

[0660] Softmax function: Calculates the probability of each emotion based on the classification results.

[0661] Device configuration and functions

[0662] The terminal provides an interface for the user to input emotions and experiences, and receives and displays feedback from the server and information on referrals to experts.

[0663] Hardware

[0664] Smartphone: A device for providing a user interface.

[0665] software

[0666] Dedicated application: An application for emotion input interface and communication with the server.

[0667] User Actions

[0668] Users access a dedicated application using their smartphone and enter their current feelings and experiences in text format, for example, "I've been feeling very sad recently. A friend of mine passed away."

[0669] The input data is sent to the server, which tokenizes the text using BertTokenizer and performs sentiment analysis using BertForSequenceClassification. Based on the analysis results, the probability of each emotion is calculated using a softmax function, and empathetic feedback is generated based on the results. For example, a message such as "I feel your emotions. I understand your sadness. I want to help you" is generated.

[0670] Based on the intensity of the emotion, referral information is also generated if necessary, and this information is presented to the user in the form of, "Serious emotions have been detected. We recommend that you consult a professional counselor."

[0671] Through the above process, the present invention can understand the user's emotions with high accuracy and provide appropriate feedback and early countermeasures.

[0672] Prompt Sentence Examples

[0673] "I've been feeling very sad lately. A friend passed away. Please analyze my current emotions and generate appropriate feedback."

[0674] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0675] Step 1:

[0676] The user uses a device to input their emotions and experiences in text form into the emotion input interface. For example, they might input, "I've been feeling very sad recently. My friend passed away." This input becomes input data for subsequent processing.

[0677] Step 2:

[0678] The terminal transmits the user's input data to the server, which includes specific operations such as packaging the data in an appropriate format and sending it to the server, for example, converting it to JSON format and sending it.

[0679] Step 3:

[0680] The server receives the user's input data and tokenizes it using BertTokenizer. Specifically, it splits the input text data into words and formats them. The output of this step is the tokenized input data.

[0681] Step 4:

[0682] The server performs sentiment analysis by inputting the tokenized input data into the BertForSequenceClassification model. This model analyzes the tokenized data and outputs the probability of each emotion (positive, negative, neutral, etc.). The output of this step is a list containing the probability of each emotion.

[0683] Step 5:

[0684] The server normalizes the resulting probability list using a softmax function and identifies the emotion with the highest probability as the emotion classification result. Specifically, given the probabilities for multiple emotions, the server selects the one with the maximum value. The output of this step is the emotion classification result (e.g., negative).

[0685] Step 6:

[0686] The server generates an empathetic feedback message based on the emotion classification result. For example, if a negative emotion is detected, it generates a message such as "I feel your emotion. I understand your sadness." The output of this step is the feedback message.

[0687] Step 7:

[0688] The server generates a referral to a specialist based on the intensity of the emotion. If a highly negative emotion is detected, it generates a referral such as "Serious emotions have been detected. We recommend that you consult a professional counselor." The output of this step is a referral to a specialist.

[0689] Step 8:

[0690] The server transmits the generated feedback message and expert referral information to the terminal, which includes specific operations of converting the data into a suitable format and transmitting it over the network.

[0691] Step 9:

[0692] The device displays the received feedback message and referral information to the expert to the user. For example, it displays a message on the screen saying, "I feel your emotions. I understand your sadness. I want to help you."

[0693] Step 10:

[0694] The user checks the displayed feedback message and information on referrals to experts, and consults with an expert if necessary.

[0695] 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.

[0696] Basic System Configuration

[0697] The system of the present invention consists of the following major components:

[0698] 1. Server

[0699] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0700] It takes input from the user and performs sentiment analysis.

[0701] Generate empathetic feedback based on analytics.

[0702] Generate specialist referrals as needed.

[0703] Use an emotion engine to identify emotions in real time in response to user input.

[0704] 2. Terminal

[0705] It provides an interface for users to input their emotions and experiences.

[0706] Send input data to the server.

[0707] View feedback and expert referrals from the server.

[0708] 3. Users

[0709] Input your emotions and experiences into the device.

[0710] Check for feedback and expert referrals from the server.

[0711] What the program does

[0712] server

[0713] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[0714] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[0715] 3. Once the user input is received, it is fed into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the input text as positive, negative, or neutral.

[0716] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[0717] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[0718] 6. Use an emotion engine to identify real-time emotions in response to user input. This emotion engine uses machine learning models to classify emotions with high accuracy, and in some cases also combines speech recognition technology to identify emotions.

[0719] Terminal

[0720] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[0721] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[0722] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[0723] user

[0724] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[0725] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[0726] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[0727] Specific examples

[0728] A specific example of the system is shown below.

[0729] Example user input

[0730] "I've been very sad recently. A friend of mine passed away."

[0731] Server response example

[0732] Sentiment analysis result: Negative

[0733] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0734] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0735] Example of terminal display

[0736] "I feel your emotions. I understand your grief. I want to help you."

[0737] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0738] This system provides users with a safe environment where they can express their emotions. They can receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide them to seek professional help.

[0739] The processing flow will be explained below.

[0740] Step 1:

[0741] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model used for natural language processing (e.g., BERT).

[0742] Step 2:

[0743] The server sets up an emotion engine, which uses machine learning models to recognize user emotions in real time.

[0744] Step 3:

[0745] The device displays an interface for users to input their emotions and experiences, providing UI elements such as input fields and buttons to allow users to easily describe their emotions.

[0746] Step 4:

[0747] The user inputs their feelings and experiences into the device. For example, the user inputs text such as, "I've been feeling very sad recently. A friend of mine passed away."

[0748] Step 5:

[0749] The terminal sends user input in real time to the server, which then passes this data to the server in the appropriate format.

[0750] Step 6:

[0751] The server inputs the received text into a sentiment analysis model to perform an initial sentiment analysis, which classifies the user's input text as positive, negative, or neutral.

[0752] Step 7:

[0753] The server uses an emotion engine to monitor changes in emotions in real time. As the user continues to input, the engine analyzes the emotions and prepares the most appropriate feedback.

[0754] Step 8:

[0755] The server generates empathetic feedback based on the final sentiment analysis result. For example, if the sentiment analysis result is negative, the server prepares feedback such as "I feel your emotion. I understand your sadness."

[0756] Step 9:

[0757] The server generates referral information for specialists based on the strength of the user's emotions. If the emotions are strong, it provides contact information for appropriate counselors and support services.

[0758] Step 10:

[0759] The server transmits the generated feedback and referral information to the expert to the terminal.

[0760] Step 11:

[0761] The terminal displays the feedback and introduction information received from the server to the user, allowing the user to receive appropriate feedback in real time.

[0762] Step 12:

[0763] Users can view the feedback displayed on their device and, if needed, seek further assistance using the provided contact details of a specialist, gaining reassurance that their feelings are understood and the immediate support they need.

[0764] Through this process, the system of the present invention analyzes the user's emotions in real time, provides empathetic feedback, and quickly refers the user to a specialist if necessary.

[0765] Example 2

[0766] 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."

[0767] Conventional emotion analysis systems have had difficulty accurately analyzing users' emotions and providing empathetic feedback based on the results. Furthermore, there was a lack of systems that could identify emotions in real time and provide referral information to specialists as needed, making it difficult to provide prompt and appropriate care for users' mental health. To solve this issue, a system capable of highly accurate emotion analysis and providing empathetic feedback in real time was needed.

[0768] 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.

[0769] In this invention, the server includes means for loading an artificial intelligence model and a tokenizer in an initialization process, means for setting up a sentiment analysis pipeline and preparing to analyze text data, means for receiving input from a user and inputting the content of the input into the sentiment analysis model, means for classifying the received input text as positive, negative, or neutral, means for generating empathetic feedback based on the analysis results, means for generating referral information to an expert if necessary depending on the strength of the emotion, and means for using an emotion engine that identifies emotions in real time, thereby enabling highly accurate analysis of user emotions, providing empathetic feedback in real time, and providing referral information to an appropriate expert if necessary.

[0770] The "initialization process" is a series of steps that loads the artificial intelligence model and tokenizer and prepares the system.

[0771] An "artificial intelligence model" is a set of pre-trained mathematical algorithms for natural language processing, including, for example, BERT.

[0772] A "tokenizer" is a tool or algorithm that breaks text into pieces that are easier to parse.

[0773] A "sentiment analysis pipeline" is the series of processing steps required to take a user's input text and analyze its sentiment.

[0774] "Text data" refers to data in the form of text that a user inputs into the system.

[0775] A "sentiment analysis model" is an algorithm that analyzes text data and classifies it as positive, negative, or neutral.

[0776] "Empathetic feedback" is a response message that is appropriate and empathetic to the user's emotional state.

[0777] "Professional Referral Information" is contact information for counselors or support services provided to users for further professional help.

[0778] "Real-time emotion identification" is the process of instantly analyzing and identifying emotions in response to user input.

[0779] An "emotion engine" is a collection of machine learning models and techniques for analyzing and identifying emotions in real time with high accuracy.

[0780] MODE FOR CARRYING OUT THE INVENTION

[0781] The system for implementing the present invention mainly comprises three components: a server, a terminal, and a user.

[0782] server

[0783] The server is responsible for central control of the system. First, the server loads the AI ​​model and tokenizer during the system initialization process. This AI model uses a pre-trained model for natural language processing, such as BERT. The tokenizer is a tool for dividing text data entered by the user into units that are easy to analyze.

[0784] Once loaded, the server sets up a sentiment analysis pipeline, ready to parse the text data from the user, using a tokenizer to convert the text into a form suitable for the model.

[0785] Upon receiving input data from the user, the server inputs the data into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the text data as positive, negative, or neutral, identifying the user's emotional state. Based on the analysis results, the server then generates an empathetic feedback message. For example, a message such as "I feel your emotions. I understand your sadness."

[0786] Depending on the strength of the emotion, it also generates referral information to a specialist. For example, if a severe negative emotion is detected, it will provide a referral such as "We recommend you consult a professional counselor." In addition, it uses an emotion engine that identifies emotions in real time, instantly analyzing and identifying the emotion in response to user input.

[0787] Terminal

[0788] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, through which users can communicate their emotions to the system. The input data is converted into an appropriate format (e.g., JSON format) and sent to the server.

[0789] The feedback received from the server and information on referrals to experts are clearly displayed on the device interface, with appropriate font sizes and colors to ensure that users can immediately understand the feedback.

[0790] user

[0791] Users input their emotions and experiences through the device interface. Specifically, they can enter detailed emotions and background information in text or voice format. After completing the input, they check the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional care for the user.

[0792] Furthermore, if the emotional state is serious, the user can check for referral information and consult with a specialist if necessary, allowing the user to receive further professional support.

[0793] Specific examples

[0794] A specific example of the system is shown below.

[0795] Example user input

[0796] "I've been very sad recently. A friend of mine passed away."

[0797] Server response example

[0798] Sentiment analysis result: Negative

[0799] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0800] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0801] Example of terminal display

[0802] "I feel your emotions. I understand your grief. I want to help you."

[0803] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0804] This system allows users to safely express their emotions and receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide users to seek professional help.

[0805] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0806] Server Processing

[0807] Step 1: Initialization process

[0808] The server loads the artificial intelligence model and tokenizer when the system starts up. Specifically, it loads a pre-trained natural language processing model such as BERT into memory and initializes the tokenizer. The input to this step is the required model and tokenizer files, and the output is the loaded model and tokenizer ready for use.

[0809] Step 2: Setting up the sentiment analysis pipeline

[0810] The server sets up a sentiment analysis pipeline, which uses a tokenizer to convert text data into tokens and format them for input to the model. For example, the tokenizer splits the text into words and subwords, which are then converted into tensors for input to the model. The input is raw text data, and the output is tokenized data in a format that the model can parse.

[0811] Step 3: Receiving User Input

[0812] The server receives the user's text data sent from the device. Once received, it parses the input data in a format such as JSON and stores it in a data structure. The input of this step is the text data entered by the user, and the output is a data structure that is ready for analysis.

[0813] Step 4: Perform sentiment analysis

[0814] The server inputs the received text data into a sentiment analysis model to classify the sentiment. Specifically, the model classifies the text as positive, negative, or neutral and calculates the probability of each category. The input is tokenized text data, and the output is the sentiment classification result (e.g., 50% positive, 40% negative, 10% neutral).

[0815] Step 5: Generate feedback

[0816] The server generates empathetic feedback based on the emotion analysis results. For example, if the emotion is strong and negative, it generates a message such as "I feel your emotion. I understand your sadness." The input is the emotion analysis results, and the output is the feedback message displayed to the user.

[0817] Step 6: Generate expert referrals

[0818] Depending on the strength of the emotion, the server generates referral information to a specialist. For example, if a severe negative emotion is detected, the server will provide a message saying, "We recommend that you consult a professional counselor," along with contact information. The input is the emotion analysis result, and the output is a message including the contact information of the specialist.

[0819] Step 7: Real-time emotion identification

[0820] The server uses an emotion engine to identify emotions in real time. When there is voice input, it uses speech recognition technology to convert the voice into text and perform emotion classification. The input is real-time voice or text data, and the output is the immediate emotion classification result.

[0821] Terminal handling

[0822] Step 1: Provide an input interface

[0823] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, which users use to input data. The input is information about the user's emotions and experiences, and the output is the input data in an organized format.

[0824] Step 2: Submitting input data

[0825] The terminal converts the user input into an appropriate format and sends it to the server. For example, it encodes text data into JSON format and sends it to the server via an HTTP request. The input is the user's input data, and the output is the data sent to the server.

[0826] Step 3: View your feedback

[0827] The feedback received from the server and the referral information to the expert are displayed to the user. Specifically, the feedback messages and referral information are displayed in an easy-to-understand interface. The input is the received feedback and referral information, and the output is the message displayed to the user.

[0828] User Action

[0829] Step 1: Input your emotions and experiences

[0830] Users input their emotions and experiences through the device interface. Specifically, they input emotions and background information in text or voice format. The input is information about the user's emotions and experiences, and the output is the data entered into the device.

[0831] Step 2: Review feedback

[0832] After completing the input, the user confirms the feedback generated by the server. The feedback is composed of empathetic and appropriate content, allowing the user to confirm it and feel that their feelings have been understood. The input is the feedback from the server, and the output is the confirmed feedback content.

[0833] Step 3: Confirm specialist referral information

[0834] In the case of a severe emotional state, referral information to a specialist is confirmed and, if necessary, a specialist is consulted, allowing the user to receive further professional assistance. The input is the referral information from the server, and the output is the confirmed specialist's contact information.

[0835] (Application example 2)

[0836] 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."

[0837] The present invention aims to provide a system that can provide both mental care and entertainment to users by not only providing appropriate feedback based on the emotions and experiences input by the user but also recommending appropriate content according to the user's emotional state.

[0838] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for recommending appropriate content based on the user's emotional state, and means for providing the content to the user. This allows the user to not only receive feedback according to their own emotional state, but also to view appropriate content.

[0839] "Sentiment analysis" is a technology that detects emotions from text data entered by the user and classifies the type of emotion (positive, negative, neutral).

[0840] "Feedback" is an empathetic message that is returned to the user based on the results of sentiment analysis, intended to understand and support the user's emotional state.

[0841] "Professional Referrals" is a way to provide links and contact information for trusted professional counselors and support services when sentiment analysis indicates a user is experiencing a serious emotional state.

[0842] "Content recommendation" is a technology that selects and provides appropriate entertainment and information such as videos and text based on the user's emotional state.

[0843] "Server" is a computer system that performs the functions of sentiment analysis, feedback generation, expert referrals, and content recommendation, and manages the overall system.

[0844] "Entertainment" refers to media such as videos, music, and games that provide users with fun and relaxation.

[0845] "User interface" refers to the means of interaction that allows users to input their emotions and experiences into the system, and to view feedback and recommended content.

[0846] "Principles of psychology" is a general term for theories and laws that provide the academic foundation for more reliable understanding and classification of emotions.

[0847] An "algorithm" is a set of procedures or computational methods for performing sentiment analysis, feedback generation, or content recommendation.

[0848] Basic System Configuration

[0849] The system of the present invention consists of the following major components:

[0850] 1. Server

[0851] The server loads AI models and tokenizers and sets up a sentiment analysis pipeline, specifically using popular models for natural language processing (e.g., BERT).

[0852] It takes input from the user and performs sentiment analysis: the server classifies the input text as positive, negative, or neutral.

[0853] Generate empathetic feedback based on the analysis, such as "I feel your emotions. I understand your sadness."

[0854] Depending on the intensity of the emotion, the system generates referral information to a specialist if necessary. For example, if a severe negative emotion is detected, the system will provide a referral such as "We recommend that you consult a professional counselor."

[0855] It recommends appropriate content based on the user's emotional state, including relaxation videos, healing music, documentaries, and more.

[0856] 2. Terminal

[0857] The device provides an interface for users to input their emotions and experiences, and through this interface users can communicate their emotions to the system.

[0858] User input is sent to the server, where it is properly formatted for sentiment analysis and feedback generation.

[0859] The system receives feedback from the server, information on referrals to experts, and recommended content, and displays it to the user. The user feels that their feelings are understood and can consult with an expert if necessary.

[0860] What the program does

[0861] server

[0862] During the initialization process, the server loads AI models and tokenizers, including natural language processing models (e.g., BERT and RoBERTa). It then sets up a sentiment analysis pipeline and prepares to analyze incoming text data. When it receives user input, it feeds the content into the sentiment analysis model to classify the sentiment. Based on the results of the sentiment analysis, it generates empathetic feedback and, if necessary, a referral to an expert. It also recommends appropriate content based on the user's emotional state.

[0863] Terminal

[0864] The device provides an interface where users can input their emotions and experiences, and sends the input to the server. The server then receives feedback, referral information to experts, and recommended content, which are then displayed to the user. This allows users to feel that their emotions are understood and to consult with experts if necessary.

[0865] user

[0866] Users enter their emotions and experiences in text format through the device interface. After completing the input, they check the feedback and recommended content generated by the server. In the case of a serious emotional state, they check the information for referrals to specialists and consult with them if necessary.

[0867] Specific examples

[0868] If a user types "I feel very tired today," the server will classify this input as "NEGATIVE" and recommend content such as healing music or meditation guides. Here are some examples of specific prompts:

[0869] python

[0870] from transformers import pipeline

[0871] emotion_pipeline = pipeline("sentiment-analysis")

[0872] User inputs emotion

[0873] user_input = "I'm very tired today"

[0874] Perform sentiment analysis

[0875] emotion = emotion_pipeline(user_input)

[0876] print(emotion)

[0877] By using this prompt, you can see the process of obtaining the sentiment analysis results.

[0878] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0879] Step 1:

[0880] The device provides an interface that allows users to input their emotions and experiences. When the user inputs something, the text data is sent to the device. The input data is text data that appropriately expresses the user's emotional state. This data is important for accurately reflecting the user's current situation.

[0881] Step 2:

[0882] The device sends the text data received from the user to the server in the appropriate format, taking care not to corrupt the data. The format is a standard data format such as JSON. This data transmission is a prerequisite for the server to accurately perform subsequent sentiment analysis.

[0883] Step 3:

[0884] The server tokenizes the received text data to input into the AI ​​model. Tokenization splits the text into words or tokens and converts it into a format that the AI ​​model can understand. This process makes it easier for the model to understand each part of the text.

[0885] Step 4:

[0886] The server inputs the tokenized data into a sentiment analysis model to classify the sentiment. This uses a generative AI model (e.g., BERT or RoBERTa) to classify sentiment as positive, negative, or neutral. This model performs highly accurate sentiment analysis based on the input data.

[0887] Step 5:

[0888] The server generates an empathetic feedback message based on the emotion classification results. For example, if a negative emotion is detected, the server generates a message saying, "I feel your emotion. I understand your sadness." This feedback helps users feel understood.

[0889] Step 6:

[0890] The server generates referral information to a specialist as needed depending on the intensity of the emotion. If a severe negative emotion is detected, the server will provide information such as "We recommend you consult a professional counselor." This allows the user to receive the necessary professional help.

[0891] Step 7:

[0892] The server then recommends content appropriate to the user's emotional state based on the results of the emotion analysis. For example, if a user inputs that they are tired, it will recommend relaxation videos or soothing music. This process is intended to provide appropriate entertainment according to the user's emotional state.

[0893] Step 8:

[0894] The server then sends the generated feedback messages, expert referrals, and recommended content to the device, formatting the data to be displayed appropriately to the user. The data is then displayed accurately on the user's device.

[0895] Step 9:

[0896] The device displays the feedback, referral information, and recommended content received from the server to the user. The user can confirm this information and feel that their feelings are understood. They can also access experts or watch recommended content as needed. This step ensures that the user receives appropriate support and entertainment from the system.

[0897] 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.

[0898] 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.

[0899] 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.

[0900] [Third embodiment]

[0901] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0902] 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.

[0903] 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).

[0904] 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.

[0905] 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.

[0906] 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).

[0907] 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.

[0908] 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.

[0909] 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.

[0910] 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.

[0911] 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.

[0912] 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."

[0913] Basic System Configuration

[0914] The system of the present invention consists of the following major components:

[0915] 1. Server

[0916] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0917] It takes input from the user and performs sentiment analysis.

[0918] Generate empathetic feedback based on analytics.

[0919] Generate specialist referrals as needed.

[0920] 2. Terminal

[0921] It provides an interface for users to input their emotions and experiences.

[0922] Send input data to the server.

[0923] View feedback and expert referrals from the server.

[0924] 3. Users

[0925] Input your emotions and experiences into the device.

[0926] Check for feedback and expert referrals from the server.

[0927] What the program does

[0928] server

[0929] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[0930] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[0931] 3. Once user input is received, it is fed into a sentiment analysis model to classify the sentiment as positive, negative, neutral, etc.

[0932] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[0933] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[0934] Terminal

[0935] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[0936] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[0937] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[0938] user

[0939] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[0940] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[0941] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[0942] Specific examples

[0943] A specific example of the system is shown below.

[0944] Example user input

[0945] "I've been very sad recently. A friend of mine passed away."

[0946] Server response example

[0947] Sentiment analysis result: Negative

[0948] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[0949] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[0950] Example of terminal display

[0951] "I feel your emotions. I understand your grief. I want to help you."

[0952] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[0953] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[0954] The processing flow will be explained below.

[0955] Step 1:

[0956] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model for natural language processing (e.g., BERT), which prepares it for sentiment analysis.

[0957] Step 2:

[0958] The device provides a user interface that allows users to input their emotions and experiences. This interface consists of a simple text input field.

[0959] Step 3:

[0960] Users input their feelings and experiences into the device. For example, a user might input, "I've been feeling very sad recently. A friend of mine passed away."

[0961] Step 4:

[0962] The terminal sends the input data from the user to the server, which then receives the data in the appropriate structured format.

[0963] Step 5:

[0964] The server inputs the received user input text into a sentiment analysis model to analyze the sentiment, which classifies the input text as positive, negative, or neutral.

[0965] Step 6:

[0966] The server generates empathetic feedback based on the analysis results. For example, if the result of the sentiment analysis is negative, it generates feedback such as, "I feel your emotion. I understand your sadness."

[0967] Step 7:

[0968] The server generates referral information based on the intensity of the emotion. If the intensity of negative emotion exceeds a certain threshold, it generates a message providing contact information for an appropriate counselor or support service.

[0969] Step 8:

[0970] The server transmits the generated feedback and expert referral information to the terminal.

[0971] Step 9:

[0972] The device displays the feedback and referral information received from the server to the user, allowing them to feel understood and identify next steps to get the professional help they need.

[0973] Step 10:

[0974] The user can then review the feedback and referral information displayed on the device and, if necessary, contact a professional counselor or support service through the provided contact details, allowing the user to take concrete action to receive appropriate assistance.

[0975] Through the above processing steps, the system of the present invention properly analyzes the user's emotions and provides empathetic feedback and referrals to necessary specialists, thereby realizing psychological care.

[0976] Example 1

[0977] 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."

[0978] Traditional sentiment analysis systems lacked the ability to properly understand users' emotions and provide empathetic feedback. Furthermore, they often lacked the ability to refer users to appropriate experts when they were in a serious emotional situation. This often resulted in insufficient emotional care for users and a dissatisfying experience.

[0979] 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.

[0980] In this invention, the server includes means for performing sentiment analysis based on emotions and experiences input by a user, means for loading a sentiment analysis model using natural language processing and setting up a sentiment analysis pipeline, means for generating empathetic feedback based on the sentiment analysis, and means for generating referral information to an expert according to the intensity of the emotion. This makes it possible to analyze a user's emotions with high accuracy, provide empathetic feedback, and, if necessary, refer the user to an expert.

[0981] "User input" refers to the user sending their emotions and experiences to the system in text form via their terminal.

[0982] "Sentiment analysis" is the process of analyzing input text data and classifying the type of emotion (positive, negative, neutral, etc.) and intensity.

[0983] "Natural language processing" is a technology that enables computers to understand, generate, and analyze human language.

[0984] A "sentiment analysis model" is an AI model trained to perform sentiment analysis, such as BERT.

[0985] A "sentiment analysis pipeline" refers to a set of processes and algorithms for analyzing text data.

[0986] "Empathetic feedback" means providing messages that empathize with and show understanding of the user's feelings.

[0987] "Information on referrals to specialists" is information that provides consultation information and contact details for appropriate specialists or counselors depending on the intensity of the user's emotions.

[0988] "Terminal" means a device that allows a user to provide input and receive feedback.

[0989] The "server" is a central computing unit that hosts the sentiment analysis model, analyzes input data from users, and generates feedback.

[0990] A "tokenizer" is a part of natural language processing that breaks down text data into a more easily processable form.

[0991] MODE FOR CARRYING OUT THE INVENTION

[0992] Basic System Configuration

[0993] The system of the present invention consists of the following main components:

[0994] 1. Server

[0995] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[0996] It takes input from the user and performs sentiment analysis.

[0997] Generate empathetic feedback based on analytics.

[0998] Generate specialist referrals as needed.

[0999] 2. Terminal

[1000] It provides an interface for users to input their emotions and experiences.

[1001] Send input data to the server.

[1002] View feedback and expert referrals from the server.

[1003] 3. Users

[1004] Input your emotions and experiences into the device.

[1005] Check for feedback and expert referrals from the server.

[1006] Specific server operations

[1007] During the initialization process, the server loads AI models for natural language processing (e.g., BERT) and tokenizers, which prepare the server to convert text data into an easily parseable format. Next, it sets up a sentiment analysis pipeline, preparing to analyze text data received from users with high accuracy.

[1008] When the server receives input from the user, it passes the content to a tokenizer, and the tokenized data is input into a sentiment analysis model. The sentiment analysis model classifies the sentiment of the text and generates an empathetic feedback message based on the results. For example, it generates a message such as, "I feel your emotion. I understand your sadness." It also generates referral information to a specialist if necessary, depending on the strength of the emotion. For example, it provides information such as, "I recommend that you consult a professional counselor."

[1009] Specific operation of the device

[1010] The device provides an interface that allows users to input their emotions and experiences. This interface includes text boxes and a submit button, allowing users to easily communicate their emotions to the system. When the user completes the input and clicks the submit button, the device sends the data to the server in an appropriate format.

[1011] The device receives feedback and referral information from the server and displays it to the user, allowing the user to feel that their feelings are understood and to obtain information on how to contact a specialist if necessary.

[1012] Specific user actions

[1013] Users input their feelings and experiences in text form through the device interface, for example, "I've been feeling very sad recently. A friend of mine passed away."

[1014] Once the user has completed and submitted the input, they will see the feedback generated by the server. This feedback is empathetic and appropriate, providing emotional support to the user. In the case of a serious emotional state, they will see information on referrals to specialists and, if necessary, consult with a specialist. This allows the user to receive further professional support.

[1015] Specific examples

[1016] A specific example of the system is shown below.

[1017] Example user input

[1018] "I've been very sad recently. A friend of mine passed away."

[1019] Server response example

[1020] Sentiment analysis result: Negative

[1021] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[1022] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[1023] Example of terminal display

[1024] "I feel your emotions. I understand your grief. I want to help you."

[1025] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[1026] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[1027] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1028] Program processing flow

[1029] Step 1: Initialize the server

[1030] How it works:

[1031] The server performs the following initial settings when the system starts up.

[1032] Input: config file and model file path.

[1033] Data processing / data computation: The server loads the AI ​​model (e.g., BERT) and tokenizer from the specified directory and sets up the sentiment analysis pipeline.

[1034] Output: The initialized sentiment analysis pipeline.

[1035] Specific behavior: Loads the model file and tokenization settings, and initializes the sentiment analysis pipeline.

[1036] Step 2: Accepting input on the terminal

[1037] How it works:

[1038] Users input their emotions and experiences through the device's interface.

[1039] Input: Text data entered by the user.

[1040] Data processing / data calculation: The terminal receives the input and retrieves the string entered in the displayed text box.

[1041] Output: User-entered text data.

[1042] Specific behavior: The device displays a text box and a submit button, and the user clicks the "Submit" button after completing the input.

[1043] Step 3: Receiving and processing text on the server

[1044] How it works:

[1045] The server receives the text data sent from the terminal and prepares it for analysis.

[1046] Input: Text data sent from the terminal.

[1047] Data processing / data operation: The server passes the received text to the tokenizer and performs tokenization.

[1048] Output: Tokenized text data.

[1049] Specific operation: The server inputs text data into the tokenizer and obtains processed tokenized data.

[1050] Step 4: Sentiment analysis on the server

[1051] How it works:

[1052] The tokenized text data is fed into a sentiment analysis model to classify sentiment.

[1053] Input: Tokenized text data.

[1054] Data processing / data calculation: Using a sentiment analysis model, perform sentiment classification (positive, negative, neutral, etc.) of text data.

[1055] Output: Sentiment analysis results.

[1056] Specific operation: The server inputs the tokenized data into the sentiment analysis model and obtains the sentiment classification result.

[1057] Step 5: Generating feedback on the server

[1058] How it works:

[1059] Generate feedback messages for users based on the results of sentiment analysis.

[1060] Input: Sentiment analysis results.

[1061] Data processing / data calculation: Based on the results of sentiment analysis, generate empathetic feedback messages and referrals to experts, if necessary.

[1062] Output: Feedback message and expert referral information.

[1063] Specific operation: The server generates a feedback message based on the emotion classification result, which contains the necessary information for the user.

[1064] Step 6: Display feedback on your device

[1065] How it works:

[1066] The feedback message and expert introduction information are received from the server and displayed to the user.

[1067] Input: Feedback message and expert referral information from the server.

[1068] Data Processing / Data Calculation: The terminal displays the entered feedback message and referral information in an appropriate format.

[1069] Output: Display of feedback message and expert referral information.

[1070] Specific operation: The device displays the data received from the server on the user interface.

[1071] Step 7: Review user feedback and act

[1072] How it works:

[1073] Users can view the feedback displayed on their device and take action if necessary.

[1074] Input: Feedback message and expert referral information.

[1075] Data processing / data calculation: The user reads the feedback and considers any necessary actions.

[1076] Output: Action decision.

[1077] Specific Action: The user reads the displayed feedback message and contacts an expert if necessary.

[1078] Through this series of processes, the system is able to analyze the user's emotions with high accuracy and provide appropriate feedback and support information.

[1079] (Application example 1)

[1080] 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."

[1081] Conventional emotion analysis systems only analyze a user's emotional state, but do not necessarily provide sufficient feedback for subsequent responses. Furthermore, they lack a mechanism for quickly taking appropriate measures when a user is experiencing stress or anxiety. In particular, when a user is experiencing high levels of stress or anxiety, it is necessary to recognize the situation early and provide appropriate support.

[1082] 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.

[1083] In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for providing an emotion input interface, and means for issuing a warning if the user's emotional state is high in stress or anxiety based on the analysis results. This makes it possible to analyze the emotions and experiences input by the user with high accuracy and provide appropriate feedback and take early measures.

[1084] "User" refers to a person who uses the system.

[1085] An "emotion input interface" is an interface that allows users to input their own emotions and experiences in text format.

[1086] "Sentiment analysis" is the process of analyzing text entered by a user and identifying the type of sentiment (positive, negative, neutral) from its content.

[1087] "Empathetic feedback" refers to messages that understand the user's emotions based on the results of sentiment analysis and respond in a way that is sympathetic to those emotions.

[1088] "Professional Referrals" refers to information that connects users to appropriate professional counselors and support services when a user is deemed to be experiencing a serious emotional state.

[1089] The "means for issuing an alert" is a means for notifying the user or other relevant parties when the user's emotional state is one of high stress or anxiety.

[1090] "Server" refers to the computer system that performs core processing such as sentiment analysis and feedback generation.

[1091] The system configuration for implementing this invention is mainly divided into two main components: a server and a terminal.

[1092] Server configuration and functions

[1093] The server analyzes the text data entered by the user based on their emotions and experiences, and generates empathetic feedback based on the results. It also provides referral information to specialists as needed. Specifically, the system uses the following hardware and software:

[1094] Hardware

[1095] Computer system: Performs core processing for sentiment analysis and feedback generation.

[1096] software

[1097] BertTokenizer: Tokenizes user-entered text and converts it into a format suitable for analysis models.

[1098] BertForSequenceClassification: Performs sentiment analysis based on tokenized input and outputs sentiment classification results.

[1099] Softmax function: Calculates the probability of each emotion based on the classification results.

[1100] Device configuration and functions

[1101] The terminal provides an interface for the user to input emotions and experiences, and receives and displays feedback from the server and information on referrals to experts.

[1102] Hardware

[1103] Smartphone: A device for providing a user interface.

[1104] software

[1105] Dedicated application: An application for emotion input interface and communication with the server.

[1106] User Actions

[1107] Users access a dedicated application using their smartphone and enter their current feelings and experiences in text format, for example, "I've been feeling very sad recently. A friend of mine passed away."

[1108] The input data is sent to the server, which tokenizes the text using BertTokenizer and performs sentiment analysis using BertForSequenceClassification. Based on the analysis results, the probability of each emotion is calculated using a softmax function, and empathetic feedback is generated based on the results. For example, a message such as "I feel your emotions. I understand your sadness. I want to help you" is generated.

[1109] Based on the intensity of the emotion, referral information is also generated if necessary, and this information is presented to the user in the form of, "Serious emotions have been detected. We recommend that you consult a professional counselor."

[1110] Through the above process, the present invention can understand the user's emotions with high accuracy and provide appropriate feedback and early countermeasures.

[1111] Prompt Sentence Examples

[1112] "I've been feeling very sad lately. A friend passed away. Please analyze my current emotions and generate appropriate feedback."

[1113] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1114] Step 1:

[1115] The user uses a device to input their emotions and experiences in text form into the emotion input interface. For example, they might input, "I've been feeling very sad recently. My friend passed away." This input becomes input data for subsequent processing.

[1116] Step 2:

[1117] The terminal transmits the user's input data to the server, which includes specific operations such as packaging the data in an appropriate format and sending it to the server, for example, converting it to JSON format and sending it.

[1118] Step 3:

[1119] The server receives the user's input data and tokenizes it using BertTokenizer. Specifically, it splits the input text data into words and formats them. The output of this step is the tokenized input data.

[1120] Step 4:

[1121] The server performs sentiment analysis by inputting the tokenized input data into the BertForSequenceClassification model. This model analyzes the tokenized data and outputs the probability of each emotion (positive, negative, neutral, etc.). The output of this step is a list containing the probability of each emotion.

[1122] Step 5:

[1123] The server normalizes the resulting probability list using a softmax function and identifies the emotion with the highest probability as the emotion classification result. Specifically, given the probabilities for multiple emotions, the server selects the one with the maximum value. The output of this step is the emotion classification result (e.g., negative).

[1124] Step 6:

[1125] The server generates an empathetic feedback message based on the emotion classification result. For example, if a negative emotion is detected, it generates a message such as "I feel your emotion. I understand your sadness." The output of this step is the feedback message.

[1126] Step 7:

[1127] The server generates a referral to a specialist based on the intensity of the emotion. If a highly negative emotion is detected, it generates a referral such as "Serious emotions have been detected. We recommend that you consult a professional counselor." The output of this step is a referral to a specialist.

[1128] Step 8:

[1129] The server transmits the generated feedback message and expert referral information to the terminal, which includes specific operations of converting the data into a suitable format and transmitting it over the network.

[1130] Step 9:

[1131] The device displays the received feedback message and referral information to the expert to the user. For example, it displays a message on the screen saying, "I feel your emotions. I understand your sadness. I want to help you."

[1132] Step 10:

[1133] The user checks the displayed feedback message and information on referrals to experts, and consults with an expert if necessary.

[1134] 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.

[1135] Basic System Configuration

[1136] The system of the present invention consists of the following major components:

[1137] 1. Server

[1138] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[1139] It takes input from the user and performs sentiment analysis.

[1140] Generate empathetic feedback based on analytics.

[1141] Generate specialist referrals as needed.

[1142] Use an emotion engine to identify emotions in real time in response to user input.

[1143] 2. Terminal

[1144] It provides an interface for users to input their emotions and experiences.

[1145] Send input data to the server.

[1146] View feedback and expert referrals from the server.

[1147] 3. Users

[1148] Input your emotions and experiences into the device.

[1149] Check for feedback and expert referrals from the server.

[1150] What the program does

[1151] server

[1152] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[1153] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[1154] 3. Once the user input is received, it is fed into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the input text as positive, negative, or neutral.

[1155] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[1156] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[1157] 6. Use an emotion engine to identify real-time emotions in response to user input. This emotion engine uses machine learning models to classify emotions with high accuracy, and in some cases also combines speech recognition technology to identify emotions.

[1158] Terminal

[1159] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[1160] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[1161] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[1162] user

[1163] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[1164] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[1165] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[1166] Specific examples

[1167] A specific example of the system is shown below.

[1168] Example user input

[1169] "I've been very sad recently. A friend of mine passed away."

[1170] Server response example

[1171] Sentiment analysis result: Negative

[1172] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[1173] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[1174] Example of terminal display

[1175] "I feel your emotions. I understand your grief. I want to help you."

[1176] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[1177] This system provides users with a safe environment where they can express their emotions. They can receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide them to seek professional help.

[1178] The processing flow will be explained below.

[1179] Step 1:

[1180] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model used for natural language processing (e.g., BERT).

[1181] Step 2:

[1182] The server sets up an emotion engine, which uses machine learning models to recognize user emotions in real time.

[1183] Step 3:

[1184] The device displays an interface for users to input their emotions and experiences, providing UI elements such as input fields and buttons to allow users to easily describe their emotions.

[1185] Step 4:

[1186] The user inputs their feelings and experiences into the device. For example, the user inputs text such as, "I've been feeling very sad recently. A friend of mine passed away."

[1187] Step 5:

[1188] The terminal sends user input in real time to the server, which then passes this data to the server in the appropriate format.

[1189] Step 6:

[1190] The server inputs the received text into a sentiment analysis model to perform an initial sentiment analysis, which classifies the user's input text as positive, negative, or neutral.

[1191] Step 7:

[1192] The server uses an emotion engine to monitor changes in emotions in real time. As the user continues to input, the engine analyzes the emotions and prepares the most appropriate feedback.

[1193] Step 8:

[1194] The server generates empathetic feedback based on the final sentiment analysis result. For example, if the sentiment analysis result is negative, the server prepares feedback such as "I feel your emotion. I understand your sadness."

[1195] Step 9:

[1196] The server generates referral information for specialists based on the strength of the user's emotions. If the emotions are strong, it provides contact information for appropriate counselors and support services.

[1197] Step 10:

[1198] The server transmits the generated feedback and referral information to the expert to the terminal.

[1199] Step 11:

[1200] The terminal displays the feedback and introduction information received from the server to the user, allowing the user to receive appropriate feedback in real time.

[1201] Step 12:

[1202] Users can view the feedback displayed on their device and, if needed, seek further assistance using the provided contact details of a specialist, gaining reassurance that their feelings are understood and the immediate support they need.

[1203] Through this process, the system of the present invention analyzes the user's emotions in real time, provides empathetic feedback, and quickly refers the user to a specialist if necessary.

[1204] Example 2

[1205] 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."

[1206] Conventional emotion analysis systems have had difficulty accurately analyzing users' emotions and providing empathetic feedback based on the results. Furthermore, there was a lack of systems that could identify emotions in real time and provide referral information to specialists as needed, making it difficult to provide prompt and appropriate care for users' mental health. To solve this issue, a system capable of highly accurate emotion analysis and providing empathetic feedback in real time was needed.

[1207] 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.

[1208] In this invention, the server includes means for loading an artificial intelligence model and a tokenizer in an initialization process, means for setting up a sentiment analysis pipeline and preparing to analyze text data, means for receiving input from a user and inputting the content of the input into the sentiment analysis model, means for classifying the received input text as positive, negative, or neutral, means for generating empathetic feedback based on the analysis results, means for generating referral information to an expert if necessary depending on the strength of the emotion, and means for using an emotion engine that identifies emotions in real time, thereby enabling highly accurate analysis of user emotions, providing empathetic feedback in real time, and providing referral information to an appropriate expert if necessary.

[1209] The "initialization process" is a series of steps that loads the artificial intelligence model and tokenizer and prepares the system.

[1210] An "artificial intelligence model" is a set of pre-trained mathematical algorithms for natural language processing, including, for example, BERT.

[1211] A "tokenizer" is a tool or algorithm that breaks text into pieces that are easier to parse.

[1212] A "sentiment analysis pipeline" is the series of processing steps required to take a user's input text and analyze its sentiment.

[1213] "Text data" refers to data in the form of text that a user inputs into the system.

[1214] A "sentiment analysis model" is an algorithm that analyzes text data and classifies it as positive, negative, or neutral.

[1215] "Empathetic feedback" is a response message that is appropriate and empathetic to the user's emotional state.

[1216] "Professional Referral Information" is contact information for counselors or support services provided to users for further professional help.

[1217] "Real-time emotion identification" is the process of instantly analyzing and identifying emotions in response to user input.

[1218] An "emotion engine" is a collection of machine learning models and techniques for analyzing and identifying emotions in real time with high accuracy.

[1219] MODE FOR CARRYING OUT THE INVENTION

[1220] The system for implementing the present invention mainly comprises three components: a server, a terminal, and a user.

[1221] server

[1222] The server is responsible for central control of the system. First, the server loads the AI ​​model and tokenizer during the system initialization process. This AI model uses a pre-trained model for natural language processing, such as BERT. The tokenizer is a tool for dividing text data entered by the user into units that are easy to analyze.

[1223] Once loaded, the server sets up a sentiment analysis pipeline, ready to parse the text data from the user, using a tokenizer to convert the text into a form suitable for the model.

[1224] Upon receiving input data from the user, the server inputs the data into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the text data as positive, negative, or neutral, identifying the user's emotional state. Based on the analysis results, the server then generates an empathetic feedback message. For example, a message such as "I feel your emotions. I understand your sadness."

[1225] Depending on the strength of the emotion, it also generates referral information to a specialist. For example, if a severe negative emotion is detected, it will provide a referral such as "We recommend you consult a professional counselor." In addition, it uses an emotion engine that identifies emotions in real time, instantly analyzing and identifying the emotion in response to user input.

[1226] Terminal

[1227] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, through which users can communicate their emotions to the system. The input data is converted into an appropriate format (e.g., JSON format) and sent to the server.

[1228] The feedback received from the server and information on referrals to experts are clearly displayed on the device interface, with appropriate font sizes and colors to ensure that users can immediately understand the feedback.

[1229] user

[1230] Users input their emotions and experiences through the device interface. Specifically, they can enter detailed emotions and background information in text or voice format. After completing the input, they check the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional care for the user.

[1231] Furthermore, if the emotional state is serious, the user can check for referral information and consult with a specialist if necessary, allowing the user to receive further professional support.

[1232] Specific examples

[1233] A specific example of the system is shown below.

[1234] Example user input

[1235] "I've been very sad recently. A friend of mine passed away."

[1236] Server response example

[1237] Sentiment analysis result: Negative

[1238] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[1239] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[1240] Example of terminal display

[1241] "I feel your emotions. I understand your grief. I want to help you."

[1242] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[1243] This system allows users to safely express their emotions and receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide users to seek professional help.

[1244] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1245] Server Processing

[1246] Step 1: Initialization process

[1247] The server loads the artificial intelligence model and tokenizer when the system starts up. Specifically, it loads a pre-trained natural language processing model such as BERT into memory and initializes the tokenizer. The input to this step is the required model and tokenizer files, and the output is the loaded model and tokenizer ready for use.

[1248] Step 2: Setting up the sentiment analysis pipeline

[1249] The server sets up a sentiment analysis pipeline, which uses a tokenizer to convert text data into tokens and format them for input to the model. For example, the tokenizer splits the text into words and subwords, which are then converted into tensors for input to the model. The input is raw text data, and the output is tokenized data in a format that the model can parse.

[1250] Step 3: Receiving User Input

[1251] The server receives the user's text data sent from the device. Once received, it parses the input data in a format such as JSON and stores it in a data structure. The input of this step is the text data entered by the user, and the output is a data structure that is ready for analysis.

[1252] Step 4: Perform sentiment analysis

[1253] The server inputs the received text data into a sentiment analysis model to classify the sentiment. Specifically, the model classifies the text as positive, negative, or neutral and calculates the probability of each category. The input is tokenized text data, and the output is the sentiment classification result (e.g., 50% positive, 40% negative, 10% neutral).

[1254] Step 5: Generate feedback

[1255] The server generates empathetic feedback based on the emotion analysis results. For example, if the emotion is strong and negative, it generates a message such as "I feel your emotion. I understand your sadness." The input is the emotion analysis results, and the output is the feedback message displayed to the user.

[1256] Step 6: Generate expert referrals

[1257] Depending on the strength of the emotion, the server generates referral information to a specialist. For example, if a severe negative emotion is detected, the server will provide a message saying, "We recommend that you consult a professional counselor," along with contact information. The input is the emotion analysis result, and the output is a message including the contact information of the specialist.

[1258] Step 7: Real-time emotion identification

[1259] The server uses an emotion engine to identify emotions in real time. When there is voice input, it uses speech recognition technology to convert the voice into text and perform emotion classification. The input is real-time voice or text data, and the output is the immediate emotion classification result.

[1260] Terminal handling

[1261] Step 1: Provide an input interface

[1262] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, which users use to input data. The input is information about the user's emotions and experiences, and the output is the input data in an organized format.

[1263] Step 2: Submitting input data

[1264] The terminal converts the user input into an appropriate format and sends it to the server. For example, it encodes text data into JSON format and sends it to the server via an HTTP request. The input is the user's input data, and the output is the data sent to the server.

[1265] Step 3: View your feedback

[1266] The feedback received from the server and the referral information to the expert are displayed to the user. Specifically, the feedback messages and referral information are displayed in an easy-to-understand interface. The input is the received feedback and referral information, and the output is the message displayed to the user.

[1267] User Action

[1268] Step 1: Input your emotions and experiences

[1269] Users input their emotions and experiences through the device interface. Specifically, they input emotions and background information in text or voice format. The input is information about the user's emotions and experiences, and the output is the data entered into the device.

[1270] Step 2: Review feedback

[1271] After completing the input, the user confirms the feedback generated by the server. The feedback is composed of empathetic and appropriate content, allowing the user to confirm it and feel that their feelings have been understood. The input is the feedback from the server, and the output is the confirmed feedback content.

[1272] Step 3: Confirm specialist referral information

[1273] In the case of a severe emotional state, referral information to a specialist is confirmed and, if necessary, a specialist is consulted, allowing the user to receive further professional assistance. The input is the referral information from the server, and the output is the confirmed specialist's contact information.

[1274] (Application example 2)

[1275] 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."

[1276] The present invention aims to provide a system that can provide both mental care and entertainment to users by not only providing appropriate feedback based on the emotions and experiences input by the user but also recommending appropriate content according to the user's emotional state.

[1277] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for recommending appropriate content based on the user's emotional state, and means for providing the content to the user. This allows the user to not only receive feedback according to their own emotional state, but also to view appropriate content.

[1278] "Sentiment analysis" is a technology that detects emotions from text data entered by the user and classifies the type of emotion (positive, negative, neutral).

[1279] "Feedback" is an empathetic message that is returned to the user based on the results of sentiment analysis, intended to understand and support the user's emotional state.

[1280] "Professional Referrals" is a way to provide links and contact information for trusted professional counselors and support services when sentiment analysis indicates a user is experiencing a serious emotional state.

[1281] "Content recommendation" is a technology that selects and provides appropriate entertainment and information such as videos and text based on the user's emotional state.

[1282] "Server" is a computer system that performs the functions of sentiment analysis, feedback generation, expert referrals, and content recommendation, and manages the overall system.

[1283] "Entertainment" refers to media such as videos, music, and games that provide users with fun and relaxation.

[1284] "User interface" refers to the means of interaction that allows users to input their emotions and experiences into the system, and to view feedback and recommended content.

[1285] "Principles of psychology" is a general term for theories and laws that provide the academic foundation for more reliable understanding and classification of emotions.

[1286] An "algorithm" is a set of procedures or computational methods for performing sentiment analysis, feedback generation, or content recommendation.

[1287] Basic System Configuration

[1288] The system of the present invention consists of the following major components:

[1289] 1. Server

[1290] The server loads AI models and tokenizers and sets up a sentiment analysis pipeline, specifically using popular models for natural language processing (e.g., BERT).

[1291] It takes input from the user and performs sentiment analysis: the server classifies the input text as positive, negative, or neutral.

[1292] Generate empathetic feedback based on the analysis, such as "I feel your emotions. I understand your sadness."

[1293] Depending on the intensity of the emotion, the system generates referral information to a specialist if necessary. For example, if a severe negative emotion is detected, the system will provide a referral such as "We recommend that you consult a professional counselor."

[1294] It recommends appropriate content based on the user's emotional state, including relaxation videos, healing music, documentaries, and more.

[1295] 2. Terminal

[1296] The device provides an interface for users to input their emotions and experiences, and through this interface users can communicate their emotions to the system.

[1297] User input is sent to the server, where it is properly formatted for sentiment analysis and feedback generation.

[1298] The system receives feedback from the server, information on referrals to experts, and recommended content, and displays it to the user. The user feels that their feelings are understood and can consult with an expert if necessary.

[1299] What the program does

[1300] server

[1301] During the initialization process, the server loads AI models and tokenizers, including natural language processing models (e.g., BERT and RoBERTa). It then sets up a sentiment analysis pipeline and prepares to analyze incoming text data. When it receives user input, it feeds the content into the sentiment analysis model to classify the sentiment. Based on the results of the sentiment analysis, it generates empathetic feedback and, if necessary, a referral to an expert. It also recommends appropriate content based on the user's emotional state.

[1302] Terminal

[1303] The device provides an interface where users can input their emotions and experiences, and sends the input to the server. The server then receives feedback, referral information to experts, and recommended content, which are then displayed to the user. This allows users to feel that their emotions are understood and to consult with experts if necessary.

[1304] user

[1305] Users enter their emotions and experiences in text format through the device interface. After completing the input, they check the feedback and recommended content generated by the server. In the case of a serious emotional state, they check the information for referrals to specialists and consult with them if necessary.

[1306] Specific examples

[1307] If a user types "I feel very tired today," the server will classify this input as "NEGATIVE" and recommend content such as healing music or meditation guides. Here are some examples of specific prompts:

[1308] python

[1309] from transformers import pipeline

[1310] emotion_pipeline = pipeline("sentiment-analysis")

[1311] User inputs emotion

[1312] user_input = "I'm very tired today"

[1313] Perform sentiment analysis

[1314] emotion = emotion_pipeline(user_input)

[1315] print(emotion)

[1316] By using this prompt, you can see the process of obtaining the sentiment analysis results.

[1317] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1318] Step 1:

[1319] The device provides an interface that allows users to input their emotions and experiences. When the user inputs something, the text data is sent to the device. The input data is text data that appropriately expresses the user's emotional state. This data is important for accurately reflecting the user's current situation.

[1320] Step 2:

[1321] The device sends the text data received from the user to the server in the appropriate format, taking care not to corrupt the data. The format is a standard data format such as JSON. This data transmission is a prerequisite for the server to accurately perform subsequent sentiment analysis.

[1322] Step 3:

[1323] The server tokenizes the received text data to input into the AI ​​model. Tokenization splits the text into words or tokens and converts it into a format that the AI ​​model can understand. This process makes it easier for the model to understand each part of the text.

[1324] Step 4:

[1325] The server inputs the tokenized data into a sentiment analysis model to classify the sentiment. This uses a generative AI model (e.g., BERT or RoBERTa) to classify sentiment as positive, negative, or neutral. This model performs highly accurate sentiment analysis based on the input data.

[1326] Step 5:

[1327] The server generates an empathetic feedback message based on the emotion classification results. For example, if a negative emotion is detected, the server generates a message saying, "I feel your emotion. I understand your sadness." This feedback helps users feel understood.

[1328] Step 6:

[1329] The server generates referral information to a specialist as needed depending on the intensity of the emotion. If a severe negative emotion is detected, the server will provide information such as "We recommend you consult a professional counselor." This allows the user to receive the necessary professional help.

[1330] Step 7:

[1331] The server then recommends content appropriate to the user's emotional state based on the results of the emotion analysis. For example, if a user inputs that they are tired, it will recommend relaxation videos or soothing music. This process is intended to provide appropriate entertainment according to the user's emotional state.

[1332] Step 8:

[1333] The server then sends the generated feedback messages, expert referrals, and recommended content to the device, formatting the data to be displayed appropriately to the user. The data is then displayed accurately on the user's device.

[1334] Step 9:

[1335] The device displays the feedback, referral information, and recommended content received from the server to the user. The user can confirm this information and feel that their feelings are understood. They can also access experts or watch recommended content as needed. This step ensures that the user receives appropriate support and entertainment from the system.

[1336] 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.

[1337] 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.

[1338] 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.

[1339] [Fourth embodiment]

[1340] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1341] 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.

[1342] 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).

[1343] 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.

[1344] 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.

[1345] 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).

[1346] 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.

[1347] 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.

[1348] 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.

[1349] 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.

[1350] 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.

[1351] 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.

[1352] 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."

[1353] Basic System Configuration

[1354] The system of the present invention consists of the following major components:

[1355] 1. Server

[1356] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[1357] It takes input from the user and performs sentiment analysis.

[1358] Generate empathetic feedback based on analytics.

[1359] Generate specialist referrals as needed.

[1360] 2. Terminal

[1361] It provides an interface for users to input their emotions and experiences.

[1362] Send input data to the server.

[1363] View feedback and expert referrals from the server.

[1364] 3. Users

[1365] Input your emotions and experiences into the device.

[1366] Check for feedback and expert referrals from the server.

[1367] What the program does

[1368] server

[1369] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[1370] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[1371] 3. Once user input is received, it is fed into a sentiment analysis model to classify the sentiment as positive, negative, neutral, etc.

[1372] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[1373] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[1374] Terminal

[1375] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[1376] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[1377] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[1378] user

[1379] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[1380] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[1381] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[1382] Specific examples

[1383] A specific example of the system is shown below.

[1384] Example user input

[1385] "I've been very sad recently. A friend of mine passed away."

[1386] Server response example

[1387] Sentiment analysis result: Negative

[1388] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[1389] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[1390] Example of terminal display

[1391] "I feel your emotions. I understand your grief. I want to help you."

[1392] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[1393] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[1394] The processing flow will be explained below.

[1395] Step 1:

[1396] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model for natural language processing (e.g., BERT), which prepares it for sentiment analysis.

[1397] Step 2:

[1398] The device provides a user interface that allows users to input their emotions and experiences. This interface consists of a simple text input field.

[1399] Step 3:

[1400] Users input their feelings and experiences into the device. For example, a user might input, "I've been feeling very sad recently. A friend of mine passed away."

[1401] Step 4:

[1402] The terminal sends the input data from the user to the server, which then receives the data in the appropriate structured format.

[1403] Step 5:

[1404] The server inputs the received user input text into a sentiment analysis model to analyze the sentiment, which classifies the input text as positive, negative, or neutral.

[1405] Step 6:

[1406] The server generates empathetic feedback based on the analysis results. For example, if the result of the sentiment analysis is negative, it generates feedback such as, "I feel your emotion. I understand your sadness."

[1407] Step 7:

[1408] The server generates referral information based on the intensity of the emotion. If the intensity of negative emotion exceeds a certain threshold, it generates a message providing contact information for an appropriate counselor or support service.

[1409] Step 8:

[1410] The server transmits the generated feedback and expert referral information to the terminal.

[1411] Step 9:

[1412] The device displays the feedback and referral information received from the server to the user, allowing them to feel understood and identify next steps to get the professional help they need.

[1413] Step 10:

[1414] The user can then review the feedback and referral information displayed on the device and, if necessary, contact a professional counselor or support service through the provided contact details, allowing the user to take concrete action to receive appropriate assistance.

[1415] Through the above processing steps, the system of the present invention properly analyzes the user's emotions and provides empathetic feedback and referrals to necessary specialists, thereby realizing psychological care.

[1416] Example 1

[1417] 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."

[1418] Traditional sentiment analysis systems lacked the ability to properly understand users' emotions and provide empathetic feedback. Furthermore, they often lacked the ability to refer users to appropriate experts when they were in a serious emotional situation. This often resulted in insufficient emotional care for users and a dissatisfying experience.

[1419] 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.

[1420] In this invention, the server includes means for performing sentiment analysis based on emotions and experiences input by a user, means for loading a sentiment analysis model using natural language processing and setting up a sentiment analysis pipeline, means for generating empathetic feedback based on the sentiment analysis, and means for generating referral information to an expert according to the intensity of the emotion. This makes it possible to analyze a user's emotions with high accuracy, provide empathetic feedback, and, if necessary, refer the user to an expert.

[1421] "User input" refers to the user sending their emotions and experiences to the system in text form via their terminal.

[1422] "Sentiment analysis" is the process of analyzing input text data and classifying the type of emotion (positive, negative, neutral, etc.) and intensity.

[1423] "Natural language processing" is a technology that enables computers to understand, generate, and analyze human language.

[1424] A "sentiment analysis model" is an AI model trained to perform sentiment analysis, such as BERT.

[1425] A "sentiment analysis pipeline" refers to a set of processes and algorithms for analyzing text data.

[1426] "Empathetic feedback" means providing messages that empathize with and show understanding of the user's feelings.

[1427] "Information on referrals to specialists" is information that provides consultation information and contact details for appropriate specialists or counselors depending on the intensity of the user's emotions.

[1428] "Terminal" means a device that allows a user to provide input and receive feedback.

[1429] The "server" is a central computing unit that hosts the sentiment analysis model, analyzes input data from users, and generates feedback.

[1430] A "tokenizer" is a part of natural language processing that breaks down text data into a more easily processable form.

[1431] MODE FOR CARRYING OUT THE INVENTION

[1432] Basic System Configuration

[1433] The system of the present invention consists of the following main components:

[1434] 1. Server

[1435] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[1436] It takes input from the user and performs sentiment analysis.

[1437] Generate empathetic feedback based on analytics.

[1438] Generate specialist referrals as needed.

[1439] 2. Terminal

[1440] It provides an interface for users to input their emotions and experiences.

[1441] Send input data to the server.

[1442] View feedback and expert referrals from the server.

[1443] 3. Users

[1444] Input your emotions and experiences into the device.

[1445] Check for feedback and expert referrals from the server.

[1446] Specific server operations

[1447] During the initialization process, the server loads AI models for natural language processing (e.g., BERT) and tokenizers, which prepare the server to convert text data into an easily parseable format. Next, it sets up a sentiment analysis pipeline, preparing to analyze text data received from users with high accuracy.

[1448] When the server receives input from the user, it passes the content to a tokenizer, and the tokenized data is input into a sentiment analysis model. The sentiment analysis model classifies the sentiment of the text and generates an empathetic feedback message based on the results. For example, it generates a message such as, "I feel your emotion. I understand your sadness." It also generates referral information to a specialist if necessary, depending on the strength of the emotion. For example, it provides information such as, "I recommend that you consult a professional counselor."

[1449] Specific operation of the device

[1450] The device provides an interface that allows users to input their emotions and experiences. This interface includes text boxes and a submit button, allowing users to easily communicate their emotions to the system. When the user completes the input and clicks the submit button, the device sends the data to the server in an appropriate format.

[1451] The device receives feedback and referral information from the server and displays it to the user, allowing the user to feel that their feelings are understood and to obtain information on how to contact a specialist if necessary.

[1452] Specific user actions

[1453] Users input their feelings and experiences in text form through the device interface, for example, "I've been feeling very sad recently. A friend of mine passed away."

[1454] Once the user has completed and submitted the input, they will see the feedback generated by the server. This feedback is empathetic and appropriate, providing emotional support to the user. In the case of a serious emotional state, they will see information on referrals to specialists and, if necessary, consult with a specialist. This allows the user to receive further professional support.

[1455] Specific examples

[1456] A specific example of the system is shown below.

[1457] Example user input

[1458] "I've been very sad recently. A friend of mine passed away."

[1459] Server response example

[1460] Sentiment analysis result: Negative

[1461] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[1462] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[1463] Example of terminal display

[1464] "I feel your emotions. I understand your grief. I want to help you."

[1465] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[1466] In this way, users are provided with a safe environment in which to express their feelings and receive the support they need along with emotional care.

[1467] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1468] Program processing flow

[1469] Step 1: Initialize the server

[1470] How it works:

[1471] The server performs the following initial settings when the system starts up.

[1472] Input: config file and model file path.

[1473] Data processing / data computation: The server loads the AI ​​model (e.g., BERT) and tokenizer from the specified directory and sets up the sentiment analysis pipeline.

[1474] Output: The initialized sentiment analysis pipeline.

[1475] Specific behavior: Loads the model file and tokenization settings, and initializes the sentiment analysis pipeline.

[1476] Step 2: Accepting input on the terminal

[1477] How it works:

[1478] Users input their emotions and experiences through the device's interface.

[1479] Input: Text data entered by the user.

[1480] Data processing / data calculation: The terminal receives the input and retrieves the string entered in the displayed text box.

[1481] Output: User-entered text data.

[1482] Specific behavior: The device displays a text box and a submit button, and the user clicks the "Submit" button after completing the input.

[1483] Step 3: Receiving and processing text on the server

[1484] How it works:

[1485] The server receives the text data sent from the terminal and prepares it for analysis.

[1486] Input: Text data sent from the terminal.

[1487] Data processing / data operation: The server passes the received text to the tokenizer and performs tokenization.

[1488] Output: Tokenized text data.

[1489] Specific operation: The server inputs text data into the tokenizer and obtains processed tokenized data.

[1490] Step 4: Sentiment analysis on the server

[1491] How it works:

[1492] The tokenized text data is fed into a sentiment analysis model to classify sentiment.

[1493] Input: Tokenized text data.

[1494] Data processing / data calculation: Using a sentiment analysis model, perform sentiment classification (positive, negative, neutral, etc.) of text data.

[1495] Output: Sentiment analysis results.

[1496] Specific operation: The server inputs the tokenized data into the sentiment analysis model and obtains the sentiment classification result.

[1497] Step 5: Generating feedback on the server

[1498] How it works:

[1499] Generate feedback messages for users based on the results of sentiment analysis.

[1500] Input: Sentiment analysis results.

[1501] Data processing / data calculation: Based on the results of sentiment analysis, generate empathetic feedback messages and referrals to experts, if necessary.

[1502] Output: Feedback message and expert referral information.

[1503] Specific operation: The server generates a feedback message based on the emotion classification result, which contains the necessary information for the user.

[1504] Step 6: Display feedback on your device

[1505] How it works:

[1506] The feedback message and expert introduction information are received from the server and displayed to the user.

[1507] Input: Feedback message and expert referral information from the server.

[1508] Data Processing / Data Calculation: The terminal displays the entered feedback message and referral information in an appropriate format.

[1509] Output: Display of feedback message and expert referral information.

[1510] Specific operation: The device displays the data received from the server on the user interface.

[1511] Step 7: Review user feedback and act

[1512] How it works:

[1513] Users can view the feedback displayed on their device and take action if necessary.

[1514] Input: Feedback message and expert referral information.

[1515] Data processing / data calculation: The user reads the feedback and considers any necessary actions.

[1516] Output: Action decision.

[1517] Specific Action: The user reads the displayed feedback message and contacts an expert if necessary.

[1518] Through this series of processes, the system is able to analyze the user's emotions with high accuracy and provide appropriate feedback and support information.

[1519] (Application example 1)

[1520] 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."

[1521] Conventional emotion analysis systems only analyze a user's emotional state, but do not necessarily provide sufficient feedback for subsequent responses. Furthermore, they lack a mechanism for quickly taking appropriate measures when a user is experiencing stress or anxiety. In particular, when a user is experiencing high levels of stress or anxiety, it is necessary to recognize the situation early and provide appropriate support.

[1522] 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.

[1523] In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for providing an emotion input interface, and means for issuing a warning if the user's emotional state is high in stress or anxiety based on the analysis results. This makes it possible to analyze the emotions and experiences input by the user with high accuracy and provide appropriate feedback and take early measures.

[1524] "User" refers to a person who uses the system.

[1525] An "emotion input interface" is an interface that allows users to input their own emotions and experiences in text format.

[1526] "Sentiment analysis" is the process of analyzing text entered by a user and identifying the type of sentiment (positive, negative, neutral) from its content.

[1527] "Empathetic feedback" refers to messages that understand the user's emotions based on the results of sentiment analysis and respond in a way that is sympathetic to those emotions.

[1528] "Professional Referrals" refers to information that connects users to appropriate professional counselors and support services when a user is deemed to be experiencing a serious emotional state.

[1529] The "means for issuing an alert" is a means for notifying the user or other relevant parties when the user's emotional state is one of high stress or anxiety.

[1530] "Server" refers to the computer system that performs core processing such as sentiment analysis and feedback generation.

[1531] The system configuration for implementing this invention is mainly divided into two main components: a server and a terminal.

[1532] Server configuration and functions

[1533] The server analyzes the text data entered by the user based on their emotions and experiences, and generates empathetic feedback based on the results. It also provides referral information to specialists as needed. Specifically, the system uses the following hardware and software:

[1534] Hardware

[1535] Computer system: Performs core processing for sentiment analysis and feedback generation.

[1536] software

[1537] BertTokenizer: Tokenizes user-entered text and converts it into a format suitable for analysis models.

[1538] BertForSequenceClassification: Performs sentiment analysis based on tokenized input and outputs sentiment classification results.

[1539] Softmax function: Calculates the probability of each emotion based on the classification results.

[1540] Device configuration and functions

[1541] The terminal provides an interface for the user to input emotions and experiences, and receives and displays feedback from the server and information on referrals to experts.

[1542] Hardware

[1543] Smartphone: A device for providing a user interface.

[1544] software

[1545] Dedicated application: An application for emotion input interface and communication with the server.

[1546] User Actions

[1547] Users access a dedicated application using their smartphone and enter their current feelings and experiences in text format, for example, "I've been feeling very sad recently. A friend of mine passed away."

[1548] The input data is sent to the server, which tokenizes the text using BertTokenizer and performs sentiment analysis using BertForSequenceClassification. Based on the analysis results, the probability of each emotion is calculated using a softmax function, and empathetic feedback is generated based on the results. For example, a message such as "I feel your emotions. I understand your sadness. I want to help you" is generated.

[1549] Based on the intensity of the emotion, referral information is also generated if necessary, and this information is presented to the user in the form of, "Serious emotions have been detected. We recommend that you consult a professional counselor."

[1550] Through the above process, the present invention can understand the user's emotions with high accuracy and provide appropriate feedback and early countermeasures.

[1551] Prompt Sentence Examples

[1552] "I've been feeling very sad lately. A friend passed away. Please analyze my current emotions and generate appropriate feedback."

[1553] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1554] Step 1:

[1555] The user uses a device to input their emotions and experiences in text form into the emotion input interface. For example, they might input, "I've been feeling very sad recently. My friend passed away." This input becomes input data for subsequent processing.

[1556] Step 2:

[1557] The terminal transmits the user's input data to the server, which includes specific operations such as packaging the data in an appropriate format and sending it to the server, for example, converting it to JSON format and sending it.

[1558] Step 3:

[1559] The server receives the user's input data and tokenizes it using BertTokenizer. Specifically, it splits the input text data into words and formats them. The output of this step is the tokenized input data.

[1560] Step 4:

[1561] The server performs sentiment analysis by inputting the tokenized input data into the BertForSequenceClassification model. This model analyzes the tokenized data and outputs the probability of each emotion (positive, negative, neutral, etc.). The output of this step is a list containing the probability of each emotion.

[1562] Step 5:

[1563] The server normalizes the resulting probability list using a softmax function and identifies the emotion with the highest probability as the emotion classification result. Specifically, given the probabilities for multiple emotions, the server selects the one with the maximum value. The output of this step is the emotion classification result (e.g., negative).

[1564] Step 6:

[1565] The server generates an empathetic feedback message based on the emotion classification result. For example, if a negative emotion is detected, it generates a message such as "I feel your emotion. I understand your sadness." The output of this step is the feedback message.

[1566] Step 7:

[1567] The server generates a referral to a specialist based on the intensity of the emotion. If a highly negative emotion is detected, it generates a referral such as "Serious emotions have been detected. We recommend that you consult a professional counselor." The output of this step is a referral to a specialist.

[1568] Step 8:

[1569] The server transmits the generated feedback message and expert referral information to the terminal, which includes specific operations of converting the data into a suitable format and transmitting it over the network.

[1570] Step 9:

[1571] The device displays the received feedback message and referral information to the expert to the user. For example, it displays a message on the screen saying, "I feel your emotions. I understand your sadness. I want to help you."

[1572] Step 10:

[1573] The user checks the displayed feedback message and information on referrals to experts, and consults with an expert if necessary.

[1574] 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.

[1575] Basic System Configuration

[1576] The system of the present invention consists of the following major components:

[1577] 1. Server

[1578] Load AI models and tokenizers to configure and manage sentiment analysis pipelines.

[1579] It takes input from the user and performs sentiment analysis.

[1580] Generate empathetic feedback based on analytics.

[1581] Generate specialist referrals as needed.

[1582] Use an emotion engine to identify emotions in real time in response to user input.

[1583] 2. Terminal

[1584] It provides an interface for users to input their emotions and experiences.

[1585] Send input data to the server.

[1586] View feedback and expert referrals from the server.

[1587] 3. Users

[1588] Input your emotions and experiences into the device.

[1589] Check for feedback and expert referrals from the server.

[1590] What the program does

[1591] server

[1592] 1. During the initialization process, the server loads an AI model and tokenizer. Specifically, it uses a popular model for natural language processing (e.g., BERT).

[1593] 2. Set up a sentiment analysis pipeline and prepare it to analyze the received text data, which will enable highly accurate sentiment analysis of user-entered sentences.

[1594] 3. Once the user input is received, it is fed into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the input text as positive, negative, or neutral.

[1595] 4. Based on the analysis results, appropriate empathetic feedback is generated for the user, such as a message like, "I feel your emotions. I understand your sadness."

[1596] 5. Depending on the intensity of the emotion, if necessary, a referral to a specialist may be generated. For example, if severe negative emotions are detected, a referral such as "We recommend that you consult a professional counselor" may be provided.

[1597] 6. Use an emotion engine to identify real-time emotions in response to user input. This emotion engine uses machine learning models to classify emotions with high accuracy, and in some cases also combines speech recognition technology to identify emotions.

[1598] Terminal

[1599] 1. The device provides an interface through which users can input their emotions and experiences. Through this interface, users can communicate their emotions to the system.

[1600] 2. The device sends the user's input to the server, where it is formatted appropriately for sentiment analysis and feedback generation.

[1601] 3. Receive feedback and referral information from the server and display it to the user, so that the user feels that their feelings are understood and can obtain information on how to contact a specialist if necessary.

[1602] user

[1603] 1. Users enter their feelings and experiences in text format through the device interface. Depending on the situation, they can also enter detailed feelings and background information.

[1604] 2. After completing the input, the user checks the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional support to the user.

[1605] 3. In case of serious emotional states, check the referral information and consult with a professional if necessary, so that the user can receive further professional help.

[1606] Specific examples

[1607] A specific example of the system is shown below.

[1608] Example user input

[1609] "I've been very sad recently. A friend of mine passed away."

[1610] Server response example

[1611] Sentiment analysis result: Negative

[1612] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[1613] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[1614] Example of terminal display

[1615] "I feel your emotions. I understand your grief. I want to help you."

[1616] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[1617] This system provides users with a safe environment where they can express their emotions. They can receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide them to seek professional help.

[1618] The processing flow will be explained below.

[1619] Step 1:

[1620] During the initialization process, the server loads an AI model and tokenizer, specifically a general-purpose model used for natural language processing (e.g., BERT).

[1621] Step 2:

[1622] The server sets up an emotion engine, which uses machine learning models to recognize user emotions in real time.

[1623] Step 3:

[1624] The device displays an interface for users to input their emotions and experiences, providing UI elements such as input fields and buttons to allow users to easily describe their emotions.

[1625] Step 4:

[1626] The user inputs their feelings and experiences into the device. For example, the user inputs text such as, "I've been feeling very sad recently. A friend of mine passed away."

[1627] Step 5:

[1628] The terminal sends user input in real time to the server, which then passes this data to the server in the appropriate format.

[1629] Step 6:

[1630] The server inputs the received text into a sentiment analysis model to perform an initial sentiment analysis, which classifies the user's input text as positive, negative, or neutral.

[1631] Step 7:

[1632] The server uses an emotion engine to monitor changes in emotions in real time. As the user continues to input, the engine analyzes the emotions and prepares the most appropriate feedback.

[1633] Step 8:

[1634] The server generates empathetic feedback based on the final sentiment analysis result. For example, if the sentiment analysis result is negative, the server prepares feedback such as "I feel your emotion. I understand your sadness."

[1635] Step 9:

[1636] The server generates referral information for specialists based on the strength of the user's emotions. If the emotions are strong, it provides contact information for appropriate counselors and support services.

[1637] Step 10:

[1638] The server transmits the generated feedback and referral information to the expert to the terminal.

[1639] Step 11:

[1640] The terminal displays the feedback and introduction information received from the server to the user, allowing the user to receive appropriate feedback in real time.

[1641] Step 12:

[1642] Users can view the feedback displayed on their device and, if needed, seek further assistance using the provided contact details of a specialist, gaining reassurance that their feelings are understood and the immediate support they need.

[1643] Through this process, the system of the present invention analyzes the user's emotions in real time, provides empathetic feedback, and quickly refers the user to a specialist if necessary.

[1644] Example 2

[1645] 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."

[1646] Conventional emotion analysis systems have had difficulty accurately analyzing users' emotions and providing empathetic feedback based on the results. Furthermore, there was a lack of systems that could identify emotions in real time and provide referral information to specialists as needed, making it difficult to provide prompt and appropriate care for users' mental health. To solve this issue, a system capable of highly accurate emotion analysis and providing empathetic feedback in real time was needed.

[1647] 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.

[1648] In this invention, the server includes means for loading an artificial intelligence model and a tokenizer in an initialization process, means for setting up a sentiment analysis pipeline and preparing to analyze text data, means for receiving input from a user and inputting the content of the input into the sentiment analysis model, means for classifying the received input text as positive, negative, or neutral, means for generating empathetic feedback based on the analysis results, means for generating referral information to an expert if necessary depending on the strength of the emotion, and means for using an emotion engine that identifies emotions in real time, thereby enabling highly accurate analysis of user emotions, providing empathetic feedback in real time, and providing referral information to an appropriate expert if necessary.

[1649] The "initialization process" is a series of steps that loads the artificial intelligence model and tokenizer and prepares the system.

[1650] An "artificial intelligence model" is a set of pre-trained mathematical algorithms for natural language processing, including, for example, BERT.

[1651] A "tokenizer" is a tool or algorithm that breaks text into pieces that are easier to parse.

[1652] A "sentiment analysis pipeline" is the series of processing steps required to take a user's input text and analyze its sentiment.

[1653] "Text data" refers to data in the form of text that a user inputs into the system.

[1654] A "sentiment analysis model" is an algorithm that analyzes text data and classifies it as positive, negative, or neutral.

[1655] "Empathetic feedback" is a response message that is appropriate and empathetic to the user's emotional state.

[1656] "Professional Referral Information" is contact information for counselors or support services provided to users for further professional help.

[1657] "Real-time emotion identification" is the process of instantly analyzing and identifying emotions in response to user input.

[1658] An "emotion engine" is a collection of machine learning models and techniques for analyzing and identifying emotions in real time with high accuracy.

[1659] MODE FOR CARRYING OUT THE INVENTION

[1660] The system for implementing the present invention mainly comprises three components: a server, a terminal, and a user.

[1661] server

[1662] The server is responsible for central control of the system. First, the server loads the AI ​​model and tokenizer during the system initialization process. This AI model uses a pre-trained model for natural language processing, such as BERT. The tokenizer is a tool for dividing text data entered by the user into units that are easy to analyze.

[1663] Once loaded, the server sets up a sentiment analysis pipeline, ready to parse the text data from the user, using a tokenizer to convert the text into a form suitable for the model.

[1664] Upon receiving input data from the user, the server inputs the data into a sentiment analysis model to analyze the sentiment. The sentiment analysis model classifies the text data as positive, negative, or neutral, identifying the user's emotional state. Based on the analysis results, the server then generates an empathetic feedback message. For example, a message such as "I feel your emotions. I understand your sadness."

[1665] Depending on the strength of the emotion, it also generates referral information to a specialist. For example, if a severe negative emotion is detected, it will provide a referral such as "We recommend you consult a professional counselor." In addition, it uses an emotion engine that identifies emotions in real time, instantly analyzing and identifying the emotion in response to user input.

[1666] Terminal

[1667] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, through which users can communicate their emotions to the system. The input data is converted into an appropriate format (e.g., JSON format) and sent to the server.

[1668] The feedback received from the server and information on referrals to experts are clearly displayed on the device interface, with appropriate font sizes and colors to ensure that users can immediately understand the feedback.

[1669] user

[1670] Users input their emotions and experiences through the device interface. Specifically, they can enter detailed emotions and background information in text or voice format. After completing the input, they check the feedback generated by the server. The feedback is empathetic and appropriate, providing emotional care for the user.

[1671] Furthermore, if the emotional state is serious, the user can check for referral information and consult with a specialist if necessary, allowing the user to receive further professional support.

[1672] Specific examples

[1673] A specific example of the system is shown below.

[1674] Example user input

[1675] "I've been very sad recently. A friend of mine passed away."

[1676] Server response example

[1677] Sentiment analysis result: Negative

[1678] Feedback message: "I feel your emotion. I understand your grief. I want to help you."

[1679] Professional referral information: "Your emotions have been very heavy. I encourage you to speak to a professional counselor. I provide their contact details below."

[1680] Example of terminal display

[1681] "I feel your emotions. I understand your grief. I want to help you."

[1682] "We feel your emotions are very heavy. We encourage you to speak to a professional counselor. We provide their contact details below."

[1683] This system allows users to safely express their emotions and receive the necessary support along with emotional care. Real-time emotion recognition also allows users to receive immediate and appropriate feedback, and in urgent cases, the system can quickly guide users to seek professional help.

[1684] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1685] Server Processing

[1686] Step 1: Initialization process

[1687] The server loads the artificial intelligence model and tokenizer when the system starts up. Specifically, it loads a pre-trained natural language processing model such as BERT into memory and initializes the tokenizer. The input to this step is the required model and tokenizer files, and the output is the loaded model and tokenizer ready for use.

[1688] Step 2: Setting up the sentiment analysis pipeline

[1689] The server sets up a sentiment analysis pipeline, which uses a tokenizer to convert text data into tokens and format them for input to the model. For example, the tokenizer splits the text into words and subwords, which are then converted into tensors for input to the model. The input is raw text data, and the output is tokenized data in a format that the model can parse.

[1690] Step 3: Receiving User Input

[1691] The server receives the user's text data sent from the device. Once received, it parses the input data in a format such as JSON and stores it in a data structure. The input of this step is the text data entered by the user, and the output is a data structure that is ready for analysis.

[1692] Step 4: Perform sentiment analysis

[1693] The server inputs the received text data into a sentiment analysis model to classify the sentiment. Specifically, the model classifies the text as positive, negative, or neutral and calculates the probability of each category. The input is tokenized text data, and the output is the sentiment classification result (e.g., 50% positive, 40% negative, 10% neutral).

[1694] Step 5: Generate feedback

[1695] The server generates empathetic feedback based on the emotion analysis results. For example, if the emotion is strong and negative, it generates a message such as "I feel your emotion. I understand your sadness." The input is the emotion analysis results, and the output is the feedback message displayed to the user.

[1696] Step 6: Generate expert referrals

[1697] Depending on the strength of the emotion, the server generates referral information to a specialist. For example, if a severe negative emotion is detected, the server will provide a message saying, "We recommend that you consult a professional counselor," along with contact information. The input is the emotion analysis result, and the output is a message including the contact information of the specialist.

[1698] Step 7: Real-time emotion identification

[1699] The server uses an emotion engine to identify emotions in real time. When there is voice input, it uses speech recognition technology to convert the voice into text and perform emotion classification. The input is real-time voice or text data, and the output is the immediate emotion classification result.

[1700] Terminal handling

[1701] Step 1: Provide an input interface

[1702] The device provides an interface that allows users to input their emotions and experiences. Specifically, it has a text input form and a voice input function, which users use to input data. The input is information about the user's emotions and experiences, and the output is the input data in an organized format.

[1703] Step 2: Submitting input data

[1704] The terminal converts the user input into an appropriate format and sends it to the server. For example, it encodes text data into JSON format and sends it to the server via an HTTP request. The input is the user's input data, and the output is the data sent to the server.

[1705] Step 3: View your feedback

[1706] The feedback received from the server and the referral information to the expert are displayed to the user. Specifically, the feedback messages and referral information are displayed in an easy-to-understand interface. The input is the received feedback and referral information, and the output is the message displayed to the user.

[1707] User Action

[1708] Step 1: Input your emotions and experiences

[1709] Users input their emotions and experiences through the device interface. Specifically, they input emotions and background information in text or voice format. The input is information about the user's emotions and experiences, and the output is the data entered into the device.

[1710] Step 2: Review feedback

[1711] After completing the input, the user confirms the feedback generated by the server. The feedback is composed of empathetic and appropriate content, allowing the user to confirm it and feel that their feelings have been understood. The input is the feedback from the server, and the output is the confirmed feedback content.

[1712] Step 3: Confirm specialist referral information

[1713] In the case of a severe emotional state, referral information to a specialist is confirmed and, if necessary, a specialist is consulted, allowing the user to receive further professional assistance. The input is the referral information from the server, and the output is the confirmed specialist's contact information.

[1714] (Application example 2)

[1715] 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."

[1716] The present invention aims to provide a system that can provide both mental care and entertainment to users by not only providing appropriate feedback based on the emotions and experiences input by the user but also recommending appropriate content according to the user's emotional state.

[1717] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for performing emotion analysis based on emotions and experiences input by the user, means for generating empathetic feedback based on the emotion analysis, means for referring the user to an expert as needed as a result of the feedback, means for recommending appropriate content based on the user's emotional state, and means for providing the content to the user. This allows the user to not only receive feedback according to their own emotional state, but also to view appropriate content.

[1718] "Sentiment analysis" is a technology that detects emotions from text data entered by the user and classifies the type of emotion (positive, negative, neutral).

[1719] "Feedback" is an empathetic message that is returned to the user based on the results of sentiment analysis, intended to understand and support the user's emotional state.

[1720] "Professional Referrals" is a way to provide links and contact information for trusted professional counselors and support services when sentiment analysis indicates a user is experiencing a serious emotional state.

[1721] "Content recommendation" is a technology that selects and provides appropriate entertainment and information such as videos and text based on the user's emotional state.

[1722] "Server" is a computer system that performs the functions of sentiment analysis, feedback generation, expert referrals, and content recommendation, and manages the overall system.

[1723] "Entertainment" refers to media such as videos, music, and games that provide users with fun and relaxation.

[1724] "User interface" refers to the means of interaction that allows users to input their emotions and experiences into the system, and to view feedback and recommended content.

[1725] "Principles of psychology" is a general term for theories and laws that provide the academic foundation for more reliable understanding and classification of emotions.

[1726] An "algorithm" is a set of procedures or computational methods for performing sentiment analysis, feedback generation, or content recommendation.

[1727] Basic System Configuration

[1728] The system of the present invention consists of the following major components:

[1729] 1. Server

[1730] The server loads AI models and tokenizers and sets up a sentiment analysis pipeline, specifically using popular models for natural language processing (e.g., BERT).

[1731] It takes input from the user and performs sentiment analysis: the server classifies the input text as positive, negative, or neutral.

[1732] Generate empathetic feedback based on the analysis, such as "I feel your emotions. I understand your sadness."

[1733] Depending on the intensity of the emotion, the system generates referral information to a specialist if necessary. For example, if a severe negative emotion is detected, the system will provide a referral such as "We recommend that you consult a professional counselor."

[1734] It recommends appropriate content based on the user's emotional state, including relaxation videos, healing music, documentaries, and more.

[1735] 2. Terminal

[1736] The device provides an interface for users to input their emotions and experiences, and through this interface users can communicate their emotions to the system.

[1737] User input is sent to the server, where it is properly formatted for sentiment analysis and feedback generation.

[1738] The system receives feedback from the server, information on referrals to experts, and recommended content, and displays it to the user. The user feels that their feelings are understood and can consult with an expert if necessary.

[1739] What the program does

[1740] server

[1741] During the initialization process, the server loads AI models and tokenizers, including natural language processing models (e.g., BERT and RoBERTa). It then sets up a sentiment analysis pipeline and prepares to analyze incoming text data. When it receives user input, it feeds the content into the sentiment analysis model to classify the sentiment. Based on the results of the sentiment analysis, it generates empathetic feedback and, if necessary, a referral to an expert. It also recommends appropriate content based on the user's emotional state.

[1742] Terminal

[1743] The device provides an interface where users can input their emotions and experiences, and sends the input to the server. The server then receives feedback, referral information to experts, and recommended content, which are then displayed to the user. This allows users to feel that their emotions are understood and to consult with experts if necessary.

[1744] user

[1745] Users enter their emotions and experiences in text format through the device interface. After completing the input, they check the feedback and recommended content generated by the server. In the case of a serious emotional state, they check the information for referrals to specialists and consult with them if necessary.

[1746] Specific examples

[1747] If a user types "I feel very tired today," the server will classify this input as "NEGATIVE" and recommend content such as healing music or meditation guides. Here are some examples of specific prompts:

[1748] python

[1749] from transformers import pipeline

[1750] emotion_pipeline = pipeline("sentiment-analysis")

[1751] User inputs emotion

[1752] user_input = "I'm very tired today"

[1753] Perform sentiment analysis

[1754] emotion = emotion_pipeline(user_input)

[1755] print(emotion)

[1756] By using this prompt, you can see the process of obtaining the sentiment analysis results.

[1757] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1758] Step 1:

[1759] The device provides an interface that allows users to input their emotions and experiences. When the user inputs something, the text data is sent to the device. The input data is text data that appropriately expresses the user's emotional state. This data is important for accurately reflecting the user's current situation.

[1760] Step 2:

[1761] The device sends the text data received from the user to the server in the appropriate format, taking care not to corrupt the data. The format is a standard data format such as JSON. This data transmission is a prerequisite for the server to accurately perform subsequent sentiment analysis.

[1762] Step 3:

[1763] The server tokenizes the received text data to input into the AI ​​model. Tokenization splits the text into words or tokens and converts it into a format that the AI ​​model can understand. This process makes it easier for the model to understand each part of the text.

[1764] Step 4:

[1765] The server inputs the tokenized data into a sentiment analysis model to classify the sentiment. This uses a generative AI model (e.g., BERT or RoBERTa) to classify sentiment as positive, negative, or neutral. This model performs highly accurate sentiment analysis based on the input data.

[1766] Step 5:

[1767] The server generates an empathetic feedback message based on the emotion classification results. For example, if a negative emotion is detected, the server generates a message saying, "I feel your emotion. I understand your sadness." This feedback helps users feel understood.

[1768] Step 6:

[1769] The server generates referral information to a specialist as needed depending on the intensity of the emotion. If a severe negative emotion is detected, the server will provide information such as "We recommend you consult a professional counselor." This allows the user to receive the necessary professional help.

[1770] Step 7:

[1771] The server then recommends content appropriate to the user's emotional state based on the results of the emotion analysis. For example, if a user inputs that they are tired, it will recommend relaxation videos or soothing music. This process is intended to provide appropriate entertainment according to the user's emotional state.

[1772] Step 8:

[1773] The server then sends the generated feedback messages, expert referrals, and recommended content to the device, formatting the data to be displayed appropriately to the user. The data is then displayed accurately on the user's device.

[1774] Step 9:

[1775] The device displays the feedback, referral information, and recommended content received from the server to the user. The user can confirm this information and feel that their feelings are understood. They can also access experts or watch recommended content as needed. This step ensures that the user receives appropriate support and entertainment from the system.

[1776] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1777] 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.

[1778] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1779] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1780] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1781] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1782] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1783] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1784] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1785] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1786] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1787] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1788] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1789] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1790] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1791] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1792] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1793] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1794] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1795] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1796] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1797] The following is further disclosed regarding the above embodiment.

[1798] (Claim 1)

[1799] A means of performing sentiment analysis based on user-entered emotions and experiences;

[1800] means for generating empathetic feedback based on said emotion analysis;

[1801] means for making referrals to specialists as necessary as a result of said feedback;

[1802] A system including:

[1803] (Claim 2)

[1804] 10. The system of claim 1, wherein the sentiment analysis is performed using an algorithm trained on principles of psychology.

[1805] (Claim 3)

[1806] 2. The system of claim 1, wherein the means for providing referrals to experts includes means for providing links or contact information to trusted professional counselors and support services.

[1807] "Example 1"

[1808] (Claim 1)

[1809] A means of performing sentiment analysis based on user-entered emotions and experiences;

[1810] means for loading a sentiment analysis model using natural language processing and configuring a sentiment analysis pipeline;

[1811] means for generating empathetic feedback based on said emotion analysis;

[1812] A means for generating referral information to an expert according to the intensity of the emotion;

[1813] A system including:

[1814] (Claim 2)

[1815] 10. The system of claim 1, wherein the sentiment analysis is performed using an algorithm trained on principles of psychology.

[1816] (Claim 3)

[1817] 2. The system of claim 1, wherein the means for providing referrals to experts includes means for providing links or contact information to trusted professional counselors and support services.

[1818] "Application Example 1"

[1819] (Claim 1)

[1820] A means of performing sentiment analysis based on user-entered emotions and experiences;

[1821] means for generating empathetic feedback based on said emotion analysis;

[1822] means for making referrals to specialists as necessary as a result of said feedback;

[1823] means for providing an emotion input interface;

[1824] and a means for issuing a warning if the user's emotional state is one of high stress or anxiety based on the analysis results.

[1825] A system including:

[1826] (Claim 2)

[1827] 10. The system of claim 1, wherein the sentiment analysis is performed using an algorithm trained on principles of psychology.

[1828] (Claim 3)

[1829] 2. The system of claim 1, wherein the means for providing referrals to experts includes means for providing links or contact information to trusted professional counselors and support services.

[1830] "Example 2: Combining Emotion Engines"

[1831] (Claim 1)

[1832] a means for loading an artificial intelligence model and a tokenizer during an initialization process;

[1833] A means for configuring a sentiment analysis pipeline and preparing text data for analysis;

[1834] means for receiving input from a user and inputting the input into a sentiment analysis model;

[1835] a means for classifying received input text as positive, negative, or neutral;

[1836] a means for generating empathetic feedback based on the analysis results;

[1837] A means for generating referral information to a specialist if necessary depending on the intensity of the emotion;

[1838] a means for using an emotion engine to identify emotions in real time;

[1839] A system including:

[1840] (Claim 2)

[1841] 10. The system of claim 1, wherein the sentiment analysis is performed using an algorithm trained on principles of psychology.

[1842] (Claim 3)

[1843] 2. The system of claim 1, wherein the means for providing referrals to experts includes means for providing links or contact information to trusted professional counselors and support services.

[1844] "Application example 2 when combining emotion engines"

[1845] (Claim 1)

[1846] A means of performing sentiment analysis based on user-entered emotions and experiences;

[1847] means for generating empathetic feedback based on said emotion analysis;

[1848] means for making referrals to specialists as necessary as a result of said feedback;

[1849] A means for recommending appropriate content based on a user's emotional state;

[1850] the means by which such content is provided to users;

[1851] A system including:

[1852] (Claim 2)

[1853] 10. The system of claim 1, wherein the sentiment analysis is performed using an algorithm trained on principles of psychology.

[1854] (Claim 3)

[1855] 2. The system of claim 1, wherein the means for referring to an expert includes means for providing links or contact information to trusted professional counselors or support services, and means for displaying recommended content based on the user's emotional state. [Explanation of symbols]

[1856] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of performing sentiment analysis based on user-entered emotions and experiences; means for generating empathetic feedback based on said emotion analysis; means for making referrals to specialists as necessary as a result of said feedback; A system including:

2. 10. The system of claim 1, wherein the sentiment analysis is performed using an algorithm trained on principles of psychology.

3. 2. The system of claim 1, wherein the means for providing referrals to experts includes means for providing links or contact information to trusted professional counselors and support services.

Images