system

Abstract

We provide the system. [Solution] A means for authenticating the user, A means for generating and managing user interactions using natural language processing, A means of analyzing dialogue data to evaluate the user's health and mental state, A means of saving the analysis results to a database and notifying when an anomaly is detected, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In an aging society with a low birthrate, the number of elderly people living alone is increasing, and the resulting solitary deaths and progression of dementia have become major social problems. In addition, there is a lack of methods for routinely monitoring the health and mental states of the elderly, and delays until abnormalities are detected may cause serious consequences. Therefore, there is a demand for a system that can provide daily support to the elderly and quickly detect abnormalities. [[ID=第37]]

Means for Solving the Problems

[0005] The present invention provides a system that includes means for authenticating a user, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, and means for storing the analysis results in a database and providing notifications when an anomaly is detected. This allows for continuous monitoring of the health and mental state of elderly individuals through daily interactions and enables rapid detection of anomalies. Furthermore, the system's security, reliability, and effectiveness are further enhanced by including means for encrypting and transmitting interaction data to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS when an anomaly is detected, and means for providing feedback based on the interaction logs.

[0006] User authentication is the process of identifying system users and verifying that they are legitimate users.

[0007] "Natural language processing" is a technology that enables computers to understand and process human language.

[0008] "Dialogue generation" is the process by which a system automatically creates the responses necessary for a natural conversation with a user.

[0009] "Dialogue management" is the process by which a system controls the flow of a conversation in order to facilitate efficient and effective interaction with users.

[0010] "Dialogue data" refers to data that digitally records the content of conversations exchanged between a user and a system.

[0011] "Health status assessment" is a process that analyzes and evaluates a user's physical health status based on dialogue data.

[0012] "Mental state assessment" is a process of analyzing and evaluating a user's mental health based on dialogue data.

[0013] "Analysis results" refer to evaluations and conclusions generated based on collected and analyzed data.

[0014] A "database" is an electronic system designed to store information efficiently and consistently, and to allow that information to be retrieved as needed.

[0015] Anomaly detection is the process of automatically identifying data or behavior that deviates from normal patterns.

[0016] "Notification" refers to a means of communicating warnings or information to relevant parties when an anomaly is detected.

[0017] "Encryption" is the process of transforming data based on a specific algorithm to make its contents unreadable to third parties.

[0018] "Facial recognition" is a technology that captures the features of a user's face and uses that information to verify their identity.

[0019] A "local government" refers to a local public entity that governs a specific region, and may include organizations that provide support for the elderly.

[0020] "Feedback" is the process of providing users with responses and advice from the system based on the logs of their interactions. [Brief explanation of the drawing]

[0021] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] [[ID=:28]]It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0022] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described according to the accompanying drawings.

[0023] First, the language used in the following description will be explained.

[0024] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0025] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0026] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0027] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0028] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0029] [First Embodiment]

[0030] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0031] As shown in Figure 1, the 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.

[0032] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0033] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0034] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0035] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0036] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0037] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0039] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0040] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0041] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0042] This invention provides a system that offers daily support to lonely elderly people and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[0043] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. It also includes means for encrypting interaction data and sending it to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, and means for providing feedback based on the interaction logs.

[0044] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[0045] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" The user's responses are recorded by the device and converted into text. This text data is then encrypted and sent to the server.

[0046] The server analyzes the received data and uses natural language processing to extract keywords and perform sentiment analysis. Based on the analysis results, it evaluates the user's health and mental state. The analysis results are recorded in a database.

[0047] The results stored in the database are constantly monitored by an algorithm designed to detect anomalies. If an anomaly is detected, the server automatically sends a notification to the local government, issuing a warning via email or SMS. This notification includes the username, the date and time the problem occurred, and a brief analysis.

[0048] Furthermore, the device provides feedback to the user based on the collected conversational data. For example, the device generates feedback to reconfirm the user's mental state, such as, "You seem a little down lately. Is there anything bothering you?"

[0049] As a concrete example, consider a scenario where User A interacts with the device daily. User A replies, "Yes, I slept well," and then the device asks, "What are your plans for today?" User A replies, "I don't have any particular plans for today." These responses are sent to the server, and the system evaluates User A's health as good. However, if User A shows lethargic responses a few days later, the server detects an abnormality in their mental state and sends a notification to the local government. In this way, abnormalities can be detected early, enabling appropriate intervention.

[0050] As described above, the present invention provides an effective system for preventing lonely deaths and the progression of dementia by routinely monitoring the health and mental state of lonely elderly people and detecting abnormalities early.

[0051] The following describes the processing flow.

[0052] Step 1:

[0053] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[0054] Step 2:

[0055] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[0056] Step 3:

[0057] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[0058] Step 4:

[0059] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[0060] Step 5:

[0061] The device encrypts the recorded text-based conversation data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[0062] Step 6:

[0063] The server decrypts the received encrypted data and uses natural language processing (NLP) techniques to extract keywords and perform sentiment analysis. Based on the user's responses, the server evaluates their health and mental state.

[0064] Step 7:

[0065] The server records the analysis results in a database. The stored data includes the date, time, user response, and analysis results.

[0066] Step 8:

[0067] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[0068] Step 9:

[0069] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[0070] Step 10:

[0071] The device provides feedback to the user. Based on past conversation data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything I can help you with?"

[0072] This processing step allows for effective monitoring of the health and safety of elderly people living alone, and enables a rapid response if any abnormalities are detected.

[0073] (Example 1)

[0074] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0075] In modern society, it is crucial to detect abnormalities in the health and mental state of lonely elderly individuals early and provide appropriate support. However, conventional systems rely on elderly individuals reporting abnormalities themselves, making early detection difficult. Furthermore, the collection and analysis of conversational data is insufficient, making it difficult to accurately monitor changes in individual health and mental states.

[0076] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0077] In this invention, the server includes means for the terminal to automatically start up at a specified time each morning and authenticate the user by scanning their face; means for generating and managing conversations with the user using natural language processing; means for recording the user's voice responses, converting them to text format and encrypting them; means for transmitting the encrypted text data to the server; means for the server to analyze the received data and perform keyword extraction and sentiment analysis using natural language processing; means for evaluating the user's health and mental state based on the analysis results and recording it in a database; and means for continuously monitoring using an algorithm for detecting abnormalities and notifying the local government if an abnormality is detected. This makes it possible to monitor the health and mental state of lonely elderly people on a daily basis, detect abnormalities early, and take appropriate action.

[0078] A "terminal" is an electronic device that functions as the user interface of a system and manages interactions with the user.

[0079] "User authentication" is a process of verifying a user's identity and is a security function that uses technologies such as facial recognition.

[0080] Natural language processing is a technology that enables computers to understand, interpret, and generate human language, and is used to generate and manage interactions with users.

[0081] "Speech recognition" is a technology that converts speech into text data, and is a technology for processing a user's verbal responses as written information.

[0082] "Encryption" is the process of hiding the contents of data in order to transfer it securely, and it is carried out using a specific algorithm.

[0083] A "server" is a centralized computer system that analyzes received data and stores and manages the results.

[0084] "Data analysis" is the process of processing collected data to derive useful information and insights, and it involves techniques such as natural language processing and other analytical methods.

[0085] "Sentiment analysis" is a technology that identifies and classifies emotions from text and audio data, and is used to evaluate a user's mental state.

[0086] A "database" is a system for efficiently storing, searching, and managing data, and is used to record analytical results.

[0087] Anomaly detection is a technique for identifying data that deviates from normal patterns, and is a process for determining whether there is an abnormality in one's health or mental state.

[0088] "Feedback" refers to advice and confirmations that a system provides based on its interaction with the user and analysis results, and includes messages that include reactions and suggestions to the user.

[0089] "Local government notification" is a function that contacts local government agencies when an anomaly is detected, and this is done via email or SMS.

[0090] Modes for carrying out the invention

[0091] This invention provides a system that offers daily support to lonely elderly individuals and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[0092] System Configuration

[0093] The system consists mainly of the following elements:

[0094] 1. Terminal

[0095] The terminal is an electronic device installed in the homes of elderly people to manage interactions with the user.

[0096] It automatically starts up at a designated time every morning and uses facial recognition technology to authenticate the user.

[0097] Conversations with users are conducted using a natural language processing engine (e.g., OpenAI®, GPT-3®).

[0098] Speech recognition software (e.g., Google® Cloud Speech-to-Text) is used to convert the user's voice response into text.

[0099] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption) and sent to the server using a secure protocol (e.g., HTTPS).

[0100] 2. Server

[0101] A server is a centralized computer system that analyzes received data and stores and manages the results.

[0102] We use natural language processing engines (e.g., NLTK and spaCy) to extract keywords and perform sentiment analysis to evaluate the user's health and mental state.

[0103] The analysis results are recorded in an SQL database (e.g., MySQL® or PostgreSQL) and continuously monitored by anomaly detection algorithms (e.g., machine learning models).

[0104] If an anomaly is detected, the system will notify the local government via email or SMS API (e.g., Twilio).

[0105] Specific operation of the system

[0106] 1. User Authentication

[0107] The device automatically starts up at 8 AM every morning and scans the user's face with its built-in camera. It also authenticates the user using a facial recognition algorithm (e.g., OpenCV or a common facial recognition library).

[0108] 2. Everyday Conversation Session

[0109] Upon successful authentication, the device activates a generative AI and begins a casual conversation with the user.

[0110] For example, questions such as "Did you sleep well last night?" and "What are your plans for today?" are generated.

[0111] 3. Data collection, encryption, and transmission

[0112] The device records the user's voice response and converts it into text data using speech recognition software.

[0113] The converted text data is encrypted and sent to the server using a secure protocol.

[0114] 4. Data Analysis and Anomaly Detection

[0115] The server analyzes the received data and uses natural language processing to evaluate the user's health and mental state.

[0116] The analysis results are recorded in a database and continuously monitored using an anomaly detection algorithm.

[0117] If an anomaly is detected, a notification will be sent to the local government via email or SMS.

[0118] 5. Provide feedback

[0119] The device provides appropriate feedback to the user based on the analysis results. It uses generative AI to generate the feedback content and converts it into speech using speech synthesis software (e.g., Amazon Polly or Google Text-to-Speech).

[0120] For example, feedback such as, "You seem a little down lately. Is there anything I can help you with?" might be provided.

[0121] Specific example

[0122] Example of interaction with User A:

[0123] Every morning at 8:00 AM, the terminal automatically starts up and scans user A's face to perform authentication.

[0124] Upon successful authentication, the device asks, "Did you sleep well last night?", and User A replies, "Yes, I slept very well."

[0125] This response will be converted to text, encrypted, and sent to the server.

[0126] The server analyzes the data and determines that User A's health status is good.

[0127] A few days later, if user A responds with "I'm tired and don't want to do anything," the server will use that data to detect an anomaly and send a notification to the local government.

[0128] The device provides feedback saying, "You seem a little down lately. Is there anything I can help you with?"

[0129] Thus, the present invention provides a system for routinely monitoring the health and mental state of lonely elderly people, detecting abnormalities early, and providing appropriate intervention.

[0130] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0131] Step 1:

[0132] System startup and user authentication

[0133] Input: System startup time, user's face image

[0134] Output: Authentication success / failure result

[0135] The device is set to automatically start up at a specified time every morning (e.g., 8:00 AM).

[0136] After startup, the built-in camera is used to capture an image of the user's face.

[0137] Using facial recognition technology (e.g., OpenCV or a general facial recognition library), this facial image is compared against pre-registered facial data.

[0138] If authentication is successful, a "Authentication successful" message is sent to the system, and the process proceeds to the next step. If it fails, the system will either retry or issue an alert.

[0139] Step 2:

[0140] Starting a daily conversation session

[0141] Input: Authentication success message

[0142] Output: Generated question prompt, user voice response

[0143] After successful authentication, the terminal launches a generated AI model (e.g., OpenAI GPT-3).

[0144] To begin speaking, generate a prompt sentence (e.g., "Did you sleep well last night?").

[0145] This question is output to the user as an audio message, and the user's voice response is collected.

[0146] Step 3:

[0147] Speech-to-text conversion and encryption

[0148] Input: User voice response

[0149] Output: Encrypted text data

[0150] The device records the collected voice responses using its microphone.

[0151] Use speech recognition software (e.g., Google Cloud Speech-to-Text) to convert recorded audio into text data.

[0152] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption).

[0153] Create encrypted text data and prepare to proceed to the next step.

[0154] Step 4:

[0155] Sending data to the server

[0156] Input: Encrypted text data

[0157] Output: Encrypted data sent to the server

[0158] The device sends encrypted text data to the server using a secure protocol (e.g., HTTPS).

[0159] The server checks the received data and prepares for the next analysis step.

[0160] Step 5:

[0161] Data Analysis

[0162] Input: Encrypted data sent to the server

[0163] Output: Analysis results (assessment of health and mental state)

[0164] The server decrypts the received encrypted data to obtain the text data.

[0165] We will analyze text data using a natural language processing engine (e.g., NLTK or spaCy). Specifically, we will perform keyword extraction and sentiment analysis.

[0166] Based on the analysis results, data is generated to evaluate the user's health and mental state. The evaluation results are then used in the next step.

[0167] Step 6:

[0168] Database recording of analysis results and anomaly detection.

[0169] Input: Analysis results

[0170] Output: Analysis results recorded in the database, anomaly detection notifications.

[0171] The server records the analysis results in an SQL database (e.g., MySQL or PostgreSQL).

[0172] Anomaly detection algorithms (e.g., machine learning models) are used to continuously monitor the analysis results stored in the database.

[0173] If an anomaly is detected, the notification system is automatically activated and sends notifications to relevant parties via email or SMS API (e.g., Twilio).

[0174] Step 7:

[0175] Provide feedback

[0176] Input: Analysis results and evaluation of anomaly detection

[0177] Output: Feedback provided to the user

[0178] The device uses a generative AI model to generate appropriate feedback (e.g., "You seem a little down lately. Is there anything I can help you with?").

[0179] Feedback generated using text-to-speech software (e.g., Amazon Polly or Google Text-to-Speech) is converted into speech and provided to the user.

[0180] The above describes the specific processing flow of this system's program.

[0181] (Application Example 1)

[0182] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0183] In modern society, monitoring the loneliness and health status of the elderly is a critical issue. In particular, the number of elderly people who are unable to receive prompt and appropriate support is increasing. Such situations can lead to serious problems such as deteriorating health and lonely deaths, necessitating effective countermeasures. Furthermore, there is a need for systems used daily by the elderly that efficiently monitor their health and mental state and respond quickly when abnormalities are detected.

[0184] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0185] In this invention, the server includes means for authenticating the user, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for performing user authentication and health checks using facial recognition technology and voice interaction, means for performing sentiment analysis of user responses using natural language processing, and means for automatically detecting anomalies and providing notifications based on the sentiment analysis results. This makes it possible to routinely monitor the health and mental state of elderly people and to quickly notify them if an anomaly is detected.

[0186] "Means of user authentication" refers to a system that uses cameras or biometric authentication technology to identify users and verify their identity.

[0187] "Means for generating and managing user interactions using natural language processing" refers to technologies that enable smooth conversations by allowing computers to understand natural language and generate and manage responses.

[0188] "Methods for analyzing dialogue data to evaluate a user's health and mental state" refers to algorithms that analyze dialogue content to diagnose and evaluate a user's health and mental state.

[0189] "A means of saving analysis results to a database and notifying when an anomaly is detected" refers to a system that stores the results of data analysis in a database and notifies when an anomaly is discovered.

[0190] "A means of performing user authentication and health checks using facial recognition technology and voice dialogue" refers to a system that utilizes facial recognition and voice dialogue technology to authenticate users and verify their health status.

[0191] "Means for performing sentiment analysis of user responses using natural language processing" refers to a technology that uses natural language processing techniques to analyze user responses and evaluate their emotional state.

[0192] "A means of automatically detecting anomalies and providing notifications based on sentiment analysis results" refers to a system that uses the results of sentiment analysis to detect anomalies and automatically provides notifications as needed.

[0193] "Regularly monitoring the user's health and mental state and promptly notifying them if an abnormality is detected" means constantly monitoring the user's health and mental state and immediately notifying relevant parties when an abnormality is discovered.

[0194] This invention is a system that routinely monitors the health and mental state of elderly individuals and promptly notifies them if any abnormalities are detected. Specific embodiments of this system are described below.

[0195] 1. System Configuration

[0196] The system consists of the following main elements:

[0197] User authentication method (facial recognition technology)

[0198] A means of generating and managing user interactions using natural language processing.

[0199] A method for analyzing dialogue data to evaluate the user's health and mental state.

[0200] A means of saving analysis results to a database and notifying when an anomaly is detected.

[0201] A method for user authentication and health checks using facial recognition technology and voice interaction.

[0202] A method for performing sentiment analysis of user responses using natural language processing.

[0203] A method for automatically detecting anomalies and sending notifications based on sentiment analysis results.

[0204] 2. Use of Hardware and Software

[0205] hardware

[0206] Camera (OpenCV compatible)

[0207] Microphone (compatible with speech_recognition library)

[0208] software

[0209] OpenCV: Uses facial recognition technology to authenticate users.

[0210] speech_recognition: Enables voice interaction and checks the user's health status.

[0211] nltk: Uses natural language processing to analyze user response text and perform sentiment analysis.

[0212] requests: Send analysis results to the remote server.

[0213] 3. Data Processing and Analysis

[0214] Dialogue data generated within the system is converted into text using speech recognition technology and further analyzed using natural language processing. During this process, sentiment analysis is performed to assess the user's health and mental state. This evaluation data is stored in a database, and if an anomaly is detected, the system automatically notifies the local government and family.

[0215] 4. Specific Examples

[0216] Example 1: User authentication and health check

[0217] The user is authenticated by facial recognition technology when they stand in front of the camera. After authentication, the system asks aloud, "Did you sleep well last night?", to which the user replies, "Yes, I slept well." This response is converted into text by speech recognition technology and sent to a server for analysis.

[0218] Example 2: Sentiment analysis and notification

[0219] If a user responds with "I don't feel like doing anything in particular today," a negative score is obtained through sentiment analysis using natural language processing. Based on this analysis, the server detects an anomaly and sends notifications to local authorities and family members.

[0220] Examples of prompts for generative AI models

[0221] Use the following text to analyze the user's health status and calculate their emotional score:

[0222] "I slept well last night."

[0223] In this way, the system can routinely monitor the health and mental state of elderly individuals and respond quickly in the event of an emergency. This enables effective interventions to prevent lonely deaths and the progression of dementia.

[0224] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0225] Step 1:

[0226] The device automatically starts up at a designated time each morning and scans the user's face using its camera. The input is the camera image, and the output is the result of user authentication. User authentication is performed using facial recognition technology (OpenCV).

[0227] Step 2:

[0228] If user authentication is successful, the device will prompt for voice input and begin checking the user's health status. It will ask questions such as, "Did you sleep well last night?" The input for this step is the user's voice response, and the output is audio data.

[0229] Step 3:

[0230] The device uses speech recognition technology (speech_recognition library) to convert the user's voice responses into text data. The input is voice data, and the output is text data.

[0231] Step 4:

[0232] Text data is sent from the terminal to the server in an encrypted state. The input here is text data, and the output is encrypted data. The requests library is used for sending.

[0233] Step 5:

[0234] The server analyzes the received text data using natural language processing (NLP) techniques (the nltk library). The input is encrypted text data, and the output is the sentiment analysis result. Specifically, it calculates a sentiment score from the text data.

[0235] Step 6:

[0236] The server evaluates the user's health and mental state based on the analysis results and saves the results to a database. The input is the emotion analysis results, and the output is the operation to save them to the database.

[0237] Step 7:

[0238] The server periodically monitors the database and sends notifications to local governments and families if an anomaly is detected. The input for this step is the analysis results from the database, and the output is a notification email or SMS.

[0239] Step 8:

[0240] The device provides feedback to the user. It generates feedback in the form of, "You seem a little down lately. Is there anything I can help you with?" The input for this step is the result of sentiment analysis, and the output is voice feedback to the user.

[0241] Step 9:

[0242] The prompt text used as input to the generative AI model is: "Analyze the user's health status using the following text and calculate a sentiment score: 'I slept well last night.'" The input is the prompt text, and the output is the generated sentiment score.

[0243] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0244] This invention relates to a system for routinely monitoring a user's health and mental state and rapidly detecting abnormalities. It is particularly characterized by its integration with an emotion engine, which allows the system to also evaluate the user's emotional state. Specific embodiments of this system are described below.

[0245] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. In addition, it includes means for encrypting and sending interaction data to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, means for providing feedback based on the interaction logs, and an emotion engine that recognizes the user's emotions.

[0246] The emotion engine analyzes audio and image data collected during interactions with the user to identify the user's emotional state. Specifically, the emotion engine uses algorithms that estimate emotions from the user's voice tone, speaking style, facial expressions, and other factors.

[0247] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[0248] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" In addition, an emotion engine analyzes the user's emotions in real time from their responses, analyzing the tone of voice and facial expressions when the user answers, "Yes, I slept well." The analysis results are converted into text format as emotion data, encrypted along with the dialogue data, and then sent to the server.

[0249] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. Specifically, both the user's responses and emotional state are considered in the health assessment. For example, even if a user answers "I slept well," if their tone of voice or facial expression is listless, they will be evaluated as mentally exhausted.

[0250] The analysis results are recorded in a database. The data includes the date, time, user response, emotional state, and analysis results. The database is constantly monitored, and an algorithm for detecting anomalies flags consecutive lethargic responses or emotional disturbances as abnormal. If an anomaly is detected, the server automatically notifies the local government via email or SMS to promptly report the user's status.

[0251] Furthermore, the device provides feedback based on the collected conversation and emotional data. For example, if a user is consistently lethargic, the device will check on the user's current situation by saying something like, "You seem a little down lately. Is there anything I can help you with?" and then provide necessary advice and support.

[0252] As a concrete example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if a mental abnormality is detected by the anomaly detection algorithm, a notification is sent to the local government. Subsequently, the local government promptly visits User B's home and takes appropriate action.

[0253] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, achieving more accurate monitoring of health and mental state. As a result, it provides an effective system that improves the quality of life for lonely elderly people and prevents lonely deaths and the progression of dementia.

[0254] The following describes the processing flow.

[0255] Step 1:

[0256] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[0257] Step 2:

[0258] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[0259] Step 3:

[0260] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[0261] Step 4:

[0262] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[0263] Step 5:

[0264] During a conversation, the device analyzes the user's voice and facial expressions using an emotion engine. It identifies the user's emotional state from factors such as voice tone, speaking style, and facial expressions.

[0265] Step 6:

[0266] The device encrypts the recorded dialogue and emotion data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[0267] Step 7:

[0268] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. The server considers both the user's responses and their emotional state when making the evaluation.

[0269] Step 8:

[0270] The server records the analysis results in a database. The stored data includes the date, time, user response, emotional state, and analysis results.

[0271] Step 9:

[0272] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[0273] Step 10:

[0274] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[0275] Step 11:

[0276] The device provides feedback to the user. Based on past conversation data and emotional data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything bothering you?"

[0277] This series of processing steps allows the system, combined with the emotion engine, to monitor the user's health and mental state with greater accuracy and respond quickly when an anomaly is detected.

[0278] (Example 2)

[0279] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0280] In modern society, health problems caused by loneliness and mental stress are becoming increasingly serious. Loneliness and mental fatigue, especially among the elderly, are significant social issues, requiring rapid and accurate monitoring and appropriate responses. However, conventional systems have limited means of evaluating users' health and mental states, and lack emotional analysis, making it difficult to accurately detect users' conditions. Furthermore, insufficient means of detecting abnormalities often led to delays in prompt responses.

[0281] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Example 2 is realized by the following means. In this invention, the server includes means for authenticating a user, means for generating and managing a dialogue with the user using natural language processing, means for analyzing dialogue data, voice data, and image data to evaluate the user's health condition and mental state, means for storing the analysis results in a database and sending a notification when an abnormality is detected, and means for identifying the user's emotional state using an emotion engine. Thereby, it becomes possible to monitor the detailed health condition and mental state including the user's emotional state, and to take prompt and appropriate actions.

[0282] "User authentication" refers to means for verifying the identity of a user when the user accesses the system using a camera or face recognition technology.

[0283] "Natural language processing" is a technology by which a computer understands, analyzes, and generates human language, and is used for means for generating and managing a dialogue with a user.

[0284] "Dialogue data" refers to data that records the content of conversations conducted between a user and a system.

[0285] "Voice data" refers to data for recording and analyzing the voice uttered by a user.

[0286] "Image data" refers to data for recording and analyzing the face and expression of a user.

[0287] "Health condition" refers to the physical and mental health condition of a user.

[0288] "Mental state" refers to an evaluation of the state of mind and emotional stability of a user.

[0289] "Analysis results" refer to evaluation results regarding the health condition and mental state of a user obtained from dialogue data and emotional data.

[0290] A "database" is a data structure used to store analysis results and interaction data so that they can be referenced and analyzed later.

[0291] "Means for notifying when an anomaly is detected" refers to a function that automatically sends alerts or notifications to relevant organizations and individuals when an anomaly is detected from the analysis results.

[0292] An "emotion engine" is a general term for algorithms and technologies used to identify a user's emotional state from their voice and facial expressions.

[0293] This invention provides a system for routinely monitoring a user's health and mental state and for rapidly detecting abnormalities. This system is implemented using the hardware and software described below.

[0294] Hardware and software:

[0295] 1. Terminal:

[0296] Camera (using facial recognition technology)

[0297] Microphone (voice data collection)

[0298] Display (for interacting with the user)

[0299] 2. Server:

[0300] Database (for storing analysis results and dialogue data)

[0301] Natural language processing models (generative AI models, e.g., GPT-3)

[0302] Emotion engine (face recognition and voice analysis)

[0303] Program processing:

[0304] User authentication:

[0305] The terminal automatically starts at the specified time every morning and scans the user's face using the camera. This authentication process uses face recognition technologies (e.g., OpenCV or FaceNet).

[0306] Daily conversation session:

[0307] If the authentication is successful, the terminal starts the generative AI model and begins a daily conversation session with the user using natural language processing technology. Specific example questions include "Did you sleep well last night?" etc.

[0308] Sentiment analysis:

[0309] During the user's response, the sentiment engine analyzes the tone of voice and expression in real time. For example, technologies such as Deep Learning for Audio are used to analyze sentiment from voice data, and OpenCV and deep learning models are used to analyze sentiment from expressions.

[0310] Data encryption and transmission:

[0311] The sentiment data and conversation data are converted into text format and encrypted using encryption technologies such as AES. The encrypted data is sent to the server.

[0312] Server analysis:

[0313] The server receives the encrypted data and performs decryption processing. Using natural language processing and the sentiment engine, it evaluates the user's health and mental state. Specifically, both the response content and the sentiment state are considered in the evaluation of the health state.

[0314] Recording in the database:

[0315] The analysis results are stored in a database, recording the date, time, response content, emotional state, and analysis results. The database is constantly monitored, and algorithms for detecting anomalies monitor for consecutive lethargic responses and emotional disturbances.

[0316] Anomaly detection and notification:

[0317] If an anomaly is detected, the server will automatically notify the local government via email or SMS. Specifically, notifications will be sent using the Twilio API, among other methods.

[0318] Provide feedback:

[0319] Based on the collected data, the device's AI model generates and provides necessary feedback to the user. This feedback includes conversational phrases such as, "You seem a little down lately. Is there anything I can help you with?"

[0320] Specific example:

[0321] For example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if the anomaly detection algorithm detects a mental abnormality, a notification is sent to the local government. The local government then promptly visits User B's home and takes appropriate action.

[0322] Examples of prompts for a generative AI model:

[0323] Enter a question such as, "Did you sleep well last night?"

[0324] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, enabling more accurate monitoring of health and mental state.

[0325] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0326] Step 1:

[0327] Device startup and user authentication

[0328] Input: Time (specified date and time)

[0329] Operation: The device will automatically start up and turn on the camera at a specified time every morning.

[0330] Data processing: The user's face is scanned using facial recognition technology (e.g., OpenCV or FaceNet) and compared with registered user data.

[0331] Output: User authentication result (success / failure)

[0332] Step 2:

[0333] Starting a daily conversation session

[0334] Input: User authentication result (success)

[0335] Operation: Upon successful authentication, the device launches a generated AI model (e.g., GPT-3) and begins a casual conversation with the user using natural language processing technology.

[0336] Prompt: Good morning. Did you sleep well last night?

[0337] Output: User response (text format)

[0338] Step 3:

[0339] User response collection and sentiment analysis

[0340] Input: User response (voice, facial expression)

[0341] Operation: The emotion engine analyzes the user's voice tone and facial expressions in real time. Deep Learning for Audio is used for audio data, and OpenCV and deep learning models are used for facial expression data.

[0342] Data processing: Convert emotional data (tone, facial expression) to text format.

[0343] Output: Sentiment data (text format)

[0344] Step 4:

[0345] Data encryption and transmission

[0346] Input: Dialogue data, emotion data (text format)

[0347] Operation: Encrypts emotional data and dialogue data using encryption technologies such as AES.

[0348] Data processing: Generation of encrypted data

[0349] Output: Encrypted data

[0350] Step 5:

[0351] Decryption of data by the server

[0352] Input: Encrypted data

[0353] Operation: The server receives the transmitted encrypted data and decrypts it using a decryption algorithm such as AES.

[0354] Data processing: Generation of decoded data

[0355] Output: Decoded data (dialogue data, sentiment data)

[0356] Step 6:

[0357] Assessment of physical and mental health

[0358] Input: Dialogue data, emotion data (decoded data)

[0359] Operation: The server uses natural language processing and an emotion engine to evaluate the user's health and mental state. Specifically, it analyzes the response using NLP technology and combines it with emotion data to perform a comprehensive evaluation.

[0360] Data processing: Generation of health status assessments

[0361] Output: Assessment results of health and mental state

[0362] Step 7:

[0363] Database recording and anomaly detection

[0364] Input: Assessment results of health and mental state

[0365] Operation: The analysis results are saved to a database. The database is constantly monitored, and an algorithm detects anomalies, monitoring for continuous lethargic responses and emotional disturbances.

[0366] Data processing: Execution of anomaly detection algorithms

[0367] Output: Anomaly detection result (normal / abnormal)

[0368] Step 8:

[0369] Notification to local government

[0370] Input: Anomaly detection result (anomaly)

[0371] Operation: If an anomaly is detected, the server will use the Twilio API or similar tools to notify the local government via email or SMS.

[0372] Data processing: Generating notification messages

[0373] Output: Notification to local government

[0374] Step 9:

[0375] Provide feedback

[0376] Input: Dialogue data, emotion data (historical data)

[0377] Operation: Based on the collected data, the device uses a generated AI model to provide necessary feedback to the user.

[0378] Prompt: You seem a little down lately. Is there anything I can help you with?

[0379] Output: Feedback content (text format)

[0380] (Application Example 2)

[0381] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0382] Conventional health and mental state monitoring systems have struggled to grasp users' emotional and health states in real time during their daily activities and to take appropriate action quickly based on that information. Furthermore, in public spaces such as stores, there have been challenges in appropriately understanding customers' emotional states and providing services based on that understanding. The present invention aims to solve these problems and provide a system that can evaluate the health and mental state of users and customers in real time and enable a rapid response when an abnormality is detected.

[0383] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0384] In this invention, the server includes means for authenticating users, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for monitoring the emotional and health states of customers in real time within the store, means for notifying store staff of the results of the customer emotion analysis and health evaluation, and means for providing customer service feedback according to the customer's state. This enables real-time monitoring of the emotional and health states of users and customers, and appropriate responses and feedback based on these.

[0385] Definitions of important words

[0386] "User authentication" is a means by which a system verifies the user's identity.

[0387] Natural language processing is a technology that enables computers to understand, generate, and manage human language.

[0388] "Dialogue data" refers to digital data that includes the content of conversations between users and systems.

[0389] "Health status" refers to information about the user's physical condition and health.

[0390] "Mental state" refers to the user's psychological or emotional state.

[0391] A "database" is a storage system for systematically saving analysis results and data.

[0392] Anomaly detection is the process of detecting unusual states or behaviors.

[0393] "Notification" refers to the act of a system informing users or administrators of specific information.

[0394] "Inside the store" refers to the physical commercial space.

[0395] "Customer" refers to an individual who uses a store or service.

[0396] Real-time monitoring is the process of collecting and analyzing information and data without delay.

[0397] "Emotional analysis" is a technology that evaluates an individual's emotional state through the analysis of voice and video.

[0398] "Health assessment results" refer to the analysis results regarding the health status of users and customers.

[0399] "Feedback" is the act of a system providing responses or advice to a user.

[0400] "Customer service" refers to the process of providing product descriptions, assistance, and customer service to customers.

[0401] Modes for carrying out the invention

[0402] This invention provides a system for routinely monitoring the health and mental state of users and customers and for rapidly detecting abnormalities. A key feature is its integration with an emotion engine, which allows for the evaluation of the emotional state of users and customers. Specific embodiments of this system are described below.

[0403] This system is used in commercial spaces (such as stores) where users and customers are present, allowing store employees to monitor customers' emotional and health states in real time using smart glasses. The main components of the system are as follows:

[0404] 1. User Authentication Method: The system uses facial recognition technology to verify the identity of users and customers. Facial recognition software will utilize libraries such as OpenCV.

[0405] 2. Natural Language Processing (NLP) Methods: The system uses NLP libraries such as Transformers and the BERT model to generate and manage conversations with users and customers. This allows for the analysis of conversation content and the evaluation of health and mental state.

[0406] 3. Emotion Analysis Methods: An emotion engine is used to analyze audio and video data and evaluate the customer's emotional state. This includes facial expression analysis and voice analysis.

[0407] 4. Database: A storage system is used to save the analyzed results. Data on customers' emotional and health states is accumulated, and an anomaly detection algorithm is applied if an anomaly is detected.

[0408] 5. Notification method: If an anomaly is detected, the server will send a notification to the store staff using a communication service such as Twilio.

[0409] 6. Real-time monitoring method: Smart glasses monitor the customer's emotional state and health status in real time and provide feedback to the store staff.

[0410] By combining the above components, it is possible to monitor the health and mental state of users and customers on a daily basis and to quickly notify them if any abnormalities are detected.

[0411] Usage example

[0412] As a concrete example, consider a scenario where a store monitors the health and emotional state of its customers. The store staff wear smart glasses. When a customer enters the store to look at merchandise, facial recognition technology identifies the customer's face, and a dialogue using natural language processing technology begins. The smart glasses analyze the customer's emotional state in real time, and data indicating that "the customer is experiencing stress" is obtained. If appropriate customer service is needed, an alert is sent to the staff via Twilio, allowing them to respond quickly and appropriately.

[0413] Example of a prompt

[0414] "Imagine a customer facial recognition system for emotion analysis. Develop an application that detects customer stress and distress, and supports appropriate feedback and responses."

[0415] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0416] Program processing steps

[0417] Step 1:

[0418] The smart glasses activate the camera and capture images of customers inside the store.

[0419] (Specific action)

[0420] The smart glasses' camera captures video in real time, and that video data is sent to the system.

[0421] (Input) Customer's real-time video data

[0422] (Output) Face feature data required for face recognition

[0423] Step 2:

[0424] The server uses facial recognition technology to identify the customer's face.

[0425] (Specific action)

[0426] The server uses the OpenCV library to perform face recognition on video data and extract customer facial features.

[0427] (Input) Facial feature data

[0428] (Output) Recognized face location information and features

[0429] Step 3:

[0430] The server activates a natural language processing model and generates and manages interactions with the user.

[0431] (Specific action)

[0432] The server uses the Transformers library and the BERT model to analyze conversations between store employees and customers in real time.

[0433] (Input) Text data of conversations with customers

[0434] (Output) Analyzed emotional and health assessment results

[0435] Step 4:

[0436] The server uses an emotion engine to analyze the customer's emotional state.

[0437] (Specific action)

[0438] The server analyzes audio and video data to assess the customer's emotional state in real time. This includes voice tone analysis and facial expression analysis.

[0439] (Input) Voice data, facial expression data

[0440] (Output) Customer's emotional state

[0441] Step 5:

[0442] The server saves the emotion analysis results to a database and detects anomalies.

[0443] (Specific action)

[0444] The server records the emotion analysis results in a database and applies an algorithm to detect consecutive abnormal emotional states.

[0445] (Input) Sentiment analysis result data

[0446] (Output) Anomaly flag and its cause data

[0447] Step 6:

[0448] The server will send a notification based on the anomaly detection results.

[0449] (Specific action)

[0450] The server uses Twilio to send notifications to store employees informing them of unusual customer behavior.

[0451] (Input) Anomaly detection flag

[0452] (Output) Notification message to store staff

[0453] Step 7:

[0454] Smart glasses provide feedback.

[0455] (Specific action)

[0456] The smart glasses display information about the customer's emotional state and hints for how to respond, providing feedback to the store staff.

[0457] (Input) Customer sentiment assessment data, health assessment data

[0458] (Output) Feedback content

[0459] Step 8:

[0460] Store staff will provide appropriate service to customers within the store.

[0461] (Specific action)

[0462] Store staff provide appropriate service and support to customers based on feedback from smart glasses.

[0463] (Input) Feedback content

[0464] (Output) Customer service and support

[0465] The above outlines the specific processing steps for implementing the invention.

[0466] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0467] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0468] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0469] [Second Embodiment]

[0470] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0471] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0472] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0473] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0474] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0475] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0476] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0477] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0478] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0479] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0480] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0481] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0482] This invention provides a system that offers daily support to lonely elderly people and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[0483] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. It also includes means for encrypting interaction data and sending it to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, and means for providing feedback based on the interaction logs.

[0484] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[0485] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" The user's responses are recorded by the device and converted into text. This text data is then encrypted and sent to the server.

[0486] The server analyzes the received data and uses natural language processing to extract keywords and perform sentiment analysis. Based on the analysis results, it evaluates the user's health and mental state. The analysis results are recorded in a database.

[0487] The results stored in the database are constantly monitored by an algorithm designed to detect anomalies. If an anomaly is detected, the server automatically sends a notification to the local government, issuing a warning via email or SMS. This notification includes the username, the date and time the problem occurred, and a brief analysis.

[0488] Furthermore, the device provides feedback to the user based on the collected conversational data. For example, the device generates feedback to reconfirm the user's mental state, such as, "You seem a little down lately. Is there anything bothering you?"

[0489] As a concrete example, consider a scenario where User A interacts with the device daily. User A replies, "Yes, I slept well," and then the device asks, "What are your plans for today?" User A replies, "I don't have any particular plans for today." These responses are sent to the server, and the system evaluates User A's health as good. However, if User A shows lethargic responses a few days later, the server detects an abnormality in their mental state and sends a notification to the local government. In this way, abnormalities can be detected early, enabling appropriate intervention.

[0490] As described above, the present invention provides an effective system for preventing lonely deaths and the progression of dementia by routinely monitoring the health and mental state of lonely elderly people and detecting abnormalities early.

[0491] The following describes the processing flow.

[0492] Step 1:

[0493] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[0494] Step 2:

[0495] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[0496] Step 3:

[0497] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[0498] Step 4:

[0499] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[0500] Step 5:

[0501] The device encrypts the recorded text-based conversation data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[0502] Step 6:

[0503] The server decrypts the received encrypted data and uses natural language processing (NLP) techniques to extract keywords and perform sentiment analysis. Based on the user's responses, the server evaluates their health and mental state.

[0504] Step 7:

[0505] The server records the analysis results in a database. The stored data includes the date, time, user response, and analysis results.

[0506] Step 8:

[0507] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[0508] Step 9:

[0509] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[0510] Step 10:

[0511] The device provides feedback to the user. Based on past conversation data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything I can help you with?"

[0512] This processing step allows for effective monitoring of the health and safety of elderly people living alone, and enables a rapid response if any abnormalities are detected.

[0513] (Example 1)

[0514] Next, we will describe Example 1. 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."

[0515] In modern society, it is crucial to detect abnormalities in the health and mental state of lonely elderly individuals early and provide appropriate support. However, conventional systems rely on elderly individuals reporting abnormalities themselves, making early detection difficult. Furthermore, the collection and analysis of conversational data is insufficient, making it difficult to accurately monitor changes in individual health and mental states.

[0516] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0517] In this invention, the server includes means for the terminal to automatically start up at a specified time each morning and authenticate the user by scanning their face; means for generating and managing conversations with the user using natural language processing; means for recording the user's voice responses, converting them to text format and encrypting them; means for transmitting the encrypted text data to the server; means for the server to analyze the received data and perform keyword extraction and sentiment analysis using natural language processing; means for evaluating the user's health and mental state based on the analysis results and recording it in a database; and means for continuously monitoring using an algorithm for detecting abnormalities and notifying the local government if an abnormality is detected. This makes it possible to monitor the health and mental state of lonely elderly people on a daily basis, detect abnormalities early, and take appropriate action.

[0518] A "terminal" is an electronic device that functions as the user interface of a system and manages interactions with the user.

[0519] "User authentication" is a process of verifying a user's identity and is a security function that uses technologies such as facial recognition.

[0520] Natural language processing is a technology that enables computers to understand, interpret, and generate human language, and is used to generate and manage interactions with users.

[0521] "Speech recognition" is a technology that converts speech into text data, and is a technology for processing a user's verbal responses as written information.

[0522] "Encryption" is the process of hiding the contents of data in order to transfer it securely, and it is carried out using a specific algorithm.

[0523] A "server" is a centralized computer system that analyzes received data and stores and manages the results.

[0524] "Data analysis" is the process of processing collected data to derive useful information and insights, and it involves techniques such as natural language processing and other analytical methods.

[0525] "Sentiment analysis" is a technology that identifies and classifies emotions from text and audio data, and is used to evaluate a user's mental state.

[0526] A "database" is a system for efficiently storing, searching, and managing data, and is used to record analytical results.

[0527] Anomaly detection is a technique for identifying data that deviates from normal patterns, and is a process for determining whether there is an abnormality in one's health or mental state.

[0528] "Feedback" refers to advice and confirmations that a system provides based on its interaction with the user and analysis results, and includes messages that include reactions and suggestions to the user.

[0529] "Local government notification" is a function that contacts local government agencies when an anomaly is detected, and this is done via email or SMS.

[0530] Modes for carrying out the invention

[0531] This invention provides a system that offers daily support to lonely elderly individuals and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[0532] System Configuration

[0533] The system consists mainly of the following elements:

[0534] 1. Terminal

[0535] The terminal is an electronic device installed in the homes of elderly people to manage interactions with the user.

[0536] It automatically starts up at a designated time every morning and uses facial recognition technology to authenticate the user.

[0537] Conversations with users are conducted using a natural language processing engine (e.g., OpenAI GPT-3).

[0538] Use speech recognition software (e.g., Google Cloud Speech-to-Text) to convert the user's voice response into text.

[0539] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption) and sent to the server using a secure protocol (e.g., HTTPS).

[0540] 2. Server

[0541] A server is a centralized computer system that analyzes received data and stores and manages the results.

[0542] We use natural language processing engines (e.g., NLTK and spaCy) to extract keywords and perform sentiment analysis to evaluate the user's health and mental state.

[0543] The analysis results are recorded in an SQL database (e.g., MySQL or PostgreSQL) and continuously monitored by anomaly detection algorithms (e.g., machine learning models).

[0544] If an anomaly is detected, the system will notify the local government via email or SMS API (e.g., Twilio).

[0545] Specific operation of the system

[0546] 1. User Authentication

[0547] The device automatically starts up at 8 AM every morning and scans the user's face with its built-in camera. It also authenticates the user using a facial recognition algorithm (e.g., OpenCV or a common facial recognition library).

[0548] 2. Everyday Conversation Session

[0549] Upon successful authentication, the device activates a generative AI and begins a casual conversation with the user.

[0550] For example, questions such as "Did you sleep well last night?" and "What are your plans for today?" are generated.

[0551] 3. Data collection, encryption, and transmission

[0552] The device records the user's voice response and converts it into text data using speech recognition software.

[0553] The converted text data is encrypted and sent to the server using a secure protocol.

[0554] 4. Data Analysis and Anomaly Detection

[0555] The server analyzes the received data and uses natural language processing to evaluate the user's health and mental state.

[0556] The analysis results are recorded in a database and continuously monitored using an anomaly detection algorithm.

[0557] If an anomaly is detected, a notification will be sent to the local government via email or SMS.

[0558] 5. Provide feedback

[0559] The device provides appropriate feedback to the user based on the analysis results. It uses generative AI to generate the feedback content and converts it into speech using speech synthesis software (e.g., Amazon Polly or Google Text-to-Speech).

[0560] For example, feedback such as, "You seem a little down lately. Is there anything I can help you with?" might be provided.

[0561] Specific example

[0562] Example of interaction with User A:

[0563] Every morning at 8:00 AM, the terminal automatically starts up and scans user A's face to perform authentication.

[0564] Upon successful authentication, the device asks, "Did you sleep well last night?", and User A replies, "Yes, I slept very well."

[0565] This response will be converted to text, encrypted, and sent to the server.

[0566] The server analyzes the data and determines that User A's health status is good.

[0567] A few days later, if user A responds with "I'm tired and don't want to do anything," the server will use that data to detect an anomaly and send a notification to the local government.

[0568] The device provides feedback saying, "You seem a little down lately. Is there anything I can help you with?"

[0569] Thus, the present invention provides a system for routinely monitoring the health and mental state of lonely elderly people, detecting abnormalities early, and providing appropriate intervention.

[0570] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0571] Step 1:

[0572] System startup and user authentication

[0573] Input: System startup time, user's face image

[0574] Output: Authentication success / failure result

[0575] The device is set to automatically start up at a specified time every morning (e.g., 8:00 AM).

[0576] After startup, the built-in camera is used to capture an image of the user's face.

[0577] Using facial recognition technology (e.g., OpenCV or a general facial recognition library), this facial image is compared against pre-registered facial data.

[0578] If authentication is successful, a "Authentication successful" message is sent to the system, and the process proceeds to the next step. If it fails, the system will either retry or issue an alert.

[0579] Step 2:

[0580] Starting a daily conversation session

[0581] Input: Authentication success message

[0582] Output: Generated question prompt, user voice response

[0583] After successful authentication, the terminal launches a generated AI model (e.g., OpenAI GPT-3).

[0584] To begin speaking, generate a prompt sentence (e.g., "Did you sleep well last night?").

[0585] This question is output to the user as an audio message, and the user's voice response is collected.

[0586] Step 3:

[0587] Speech-to-text conversion and encryption

[0588] Input: User voice response

[0589] Output: Encrypted text data

[0590] The device records the collected voice responses using its microphone.

[0591] Use speech recognition software (e.g., Google Cloud Speech-to-Text) to convert recorded audio into text data.

[0592] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption).

[0593] Create encrypted text data and prepare to proceed to the next step.

[0594] Step 4:

[0595] Sending data to the server

[0596] Input: Encrypted text data

[0597] Output: Encrypted data sent to the server

[0598] The device sends encrypted text data to the server using a secure protocol (e.g., HTTPS).

[0599] The server checks the received data and prepares for the next analysis step.

[0600] Step 5:

[0601] Data Analysis

[0602] Input: Encrypted data sent to the server

[0603] Output: Analysis results (assessment of health and mental state)

[0604] The server decrypts the received encrypted data to obtain the text data.

[0605] We will analyze text data using a natural language processing engine (e.g., NLTK or spaCy). Specifically, we will perform keyword extraction and sentiment analysis.

[0606] Based on the analysis results, data is generated to evaluate the user's health and mental state. The evaluation results are then used in the next step.

[0607] Step 6:

[0608] Database recording of analysis results and anomaly detection.

[0609] Input: Analysis results

[0610] Output: Analysis results recorded in the database, anomaly detection notifications.

[0611] The server records the analysis results in an SQL database (e.g., MySQL or PostgreSQL).

[0612] Anomaly detection algorithms (e.g., machine learning models) are used to continuously monitor the analysis results stored in the database.

[0613] If an anomaly is detected, the notification system is automatically activated and sends notifications to relevant parties via email or SMS API (e.g., Twilio).

[0614] Step 7:

[0615] Provide feedback

[0616] Input: Analysis results and evaluation of anomaly detection

[0617] Output: Feedback provided to the user

[0618] The device uses a generative AI model to generate appropriate feedback (e.g., "You seem a little down lately. Is there anything I can help you with?").

[0619] Feedback generated using text-to-speech software (e.g., Amazon Polly or Google Text-to-Speech) is converted into speech and provided to the user.

[0620] The above describes the specific processing flow of this system's program.

[0621] (Application Example 1)

[0622] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0623] In modern society, monitoring the loneliness and health status of the elderly is a critical issue. In particular, the number of elderly people who are unable to receive prompt and appropriate support is increasing. Such situations can lead to serious problems such as deteriorating health and lonely deaths, necessitating effective countermeasures. Furthermore, there is a need for systems used daily by the elderly that efficiently monitor their health and mental state and respond quickly when abnormalities are detected.

[0624] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0625] In this invention, the server includes means for authenticating the user, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for performing user authentication and health checks using facial recognition technology and voice interaction, means for performing sentiment analysis of user responses using natural language processing, and means for automatically detecting anomalies and providing notifications based on the sentiment analysis results. This makes it possible to routinely monitor the health and mental state of elderly people and to quickly notify them if an anomaly is detected.

[0626] "Means of user authentication" refers to a system that uses cameras or biometric authentication technology to identify users and verify their identity.

[0627] "Means for generating and managing user interactions using natural language processing" refers to technologies that enable smooth conversations by allowing computers to understand natural language and generate and manage responses.

[0628] "Methods for analyzing dialogue data to evaluate a user's health and mental state" refers to algorithms that analyze dialogue content to diagnose and evaluate a user's health and mental state.

[0629] "A means of saving analysis results to a database and notifying when an anomaly is detected" refers to a system that stores the results of data analysis in a database and notifies when an anomaly is discovered.

[0630] "A means of performing user authentication and health checks using facial recognition technology and voice dialogue" refers to a system that utilizes facial recognition and voice dialogue technology to authenticate users and verify their health status.

[0631] "Means for performing sentiment analysis of user responses using natural language processing" refers to a technology that uses natural language processing techniques to analyze user responses and evaluate their emotional state.

[0632] "A means of automatically detecting anomalies and providing notifications based on sentiment analysis results" refers to a system that uses the results of sentiment analysis to detect anomalies and automatically provides notifications as needed.

[0633] "Regularly monitoring the user's health and mental state and promptly notifying them if an abnormality is detected" means constantly monitoring the user's health and mental state and immediately notifying relevant parties when an abnormality is discovered.

[0634] This invention is a system that routinely monitors the health and mental state of elderly individuals and promptly notifies them if any abnormalities are detected. Specific embodiments of this system are described below.

[0635] 1. System Configuration

[0636] The system consists of the following main elements:

[0637] User authentication method (facial recognition technology)

[0638] A means of generating and managing user interactions using natural language processing.

[0639] A method for analyzing dialogue data to evaluate the user's health and mental state.

[0640] A means of saving analysis results to a database and notifying when an anomaly is detected.

[0641] A method for user authentication and health checks using facial recognition technology and voice interaction.

[0642] A method for performing sentiment analysis of user responses using natural language processing.

[0643] A method for automatically detecting anomalies and sending notifications based on sentiment analysis results.

[0644] 2. Use of Hardware and Software

[0645] hardware

[0646] Camera (OpenCV compatible)

[0647] Microphone (compatible with speech_recognition library)

[0648] software

[0649] OpenCV: Uses facial recognition technology to authenticate users.

[0650] speech_recognition: Enables voice interaction and checks the user's health status.

[0651] nltk: Uses natural language processing to analyze user response text and perform sentiment analysis.

[0652] requests: Send analysis results to the remote server.

[0653] 3. Data Processing and Analysis

[0654] Dialogue data generated within the system is converted into text using speech recognition technology and further analyzed using natural language processing. During this process, sentiment analysis is performed to assess the user's health and mental state. This evaluation data is stored in a database, and if an anomaly is detected, the system automatically notifies the local government and family.

[0655] 4. Specific Examples

[0656] Example 1: User authentication and health check

[0657] The user is authenticated by facial recognition technology when they stand in front of the camera. After authentication, the system asks aloud, "Did you sleep well last night?", to which the user replies, "Yes, I slept well." This response is converted into text by speech recognition technology and sent to a server for analysis.

[0658] Example 2: Sentiment analysis and notification

[0659] If a user responds with "I don't feel like doing anything in particular today," a negative score is obtained through sentiment analysis using natural language processing. Based on this analysis, the server detects an anomaly and sends notifications to local authorities and family members.

[0660] Examples of prompts for generative AI models

[0661] Use the following text to analyze the user's health status and calculate their emotional score:

[0662] "I slept well last night."

[0663] In this way, the system can routinely monitor the health and mental state of elderly individuals and respond quickly in the event of an emergency. This enables effective interventions to prevent lonely deaths and the progression of dementia.

[0664] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0665] Step 1:

[0666] The device automatically starts up at a designated time each morning and scans the user's face using its camera. The input is the camera image, and the output is the result of user authentication. User authentication is performed using facial recognition technology (OpenCV).

[0667] Step 2:

[0668] If user authentication is successful, the device will prompt for voice input and begin checking the user's health status. It will ask questions such as, "Did you sleep well last night?" The input for this step is the user's voice response, and the output is audio data.

[0669] Step 3:

[0670] The device uses speech recognition technology (speech_recognition library) to convert the user's voice responses into text data. The input is voice data, and the output is text data.

[0671] Step 4:

[0672] Text data is sent from the terminal to the server in an encrypted state. The input here is text data, and the output is encrypted data. The requests library is used for sending.

[0673] Step 5:

[0674] The server analyzes the received text data using natural language processing (NLP) techniques (the nltk library). The input is encrypted text data, and the output is the sentiment analysis result. Specifically, it calculates a sentiment score from the text data.

[0675] Step 6:

[0676] The server evaluates the user's health and mental state based on the analysis results and saves the results to a database. The input is the emotion analysis results, and the output is the operation to save them to the database.

[0677] Step 7:

[0678] The server periodically monitors the database and sends notifications to local governments and families if an anomaly is detected. The input for this step is the analysis results from the database, and the output is a notification email or SMS.

[0679] Step 8:

[0680] The device provides feedback to the user. It generates feedback in the form of, "You seem a little down lately. Is there anything I can help you with?" The input for this step is the result of sentiment analysis, and the output is voice feedback to the user.

[0681] Step 9:

[0682] The prompt text used as input to the generative AI model is: "Analyze the user's health status using the following text and calculate a sentiment score: 'I slept well last night.'" The input is the prompt text, and the output is the generated sentiment score.

[0683] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0684] This invention relates to a system for routinely monitoring a user's health and mental state and rapidly detecting abnormalities. It is particularly characterized by its integration with an emotion engine, which allows the system to also evaluate the user's emotional state. Specific embodiments of this system are described below.

[0685] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. In addition, it includes means for encrypting and sending interaction data to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, means for providing feedback based on the interaction logs, and an emotion engine that recognizes the user's emotions.

[0686] The emotion engine analyzes audio and image data collected during interactions with the user to identify the user's emotional state. Specifically, the emotion engine uses algorithms that estimate emotions from the user's voice tone, speaking style, facial expressions, and other factors.

[0687] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[0688] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" In addition, an emotion engine analyzes the user's emotions in real time from their responses, analyzing the tone of voice and facial expressions when the user answers, "Yes, I slept well." The analysis results are converted into text format as emotion data, encrypted along with the dialogue data, and then sent to the server.

[0689] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. Specifically, both the user's responses and emotional state are considered in the health assessment. For example, even if a user answers "I slept well," if their tone of voice or facial expression is listless, they will be evaluated as mentally exhausted.

[0690] The analysis results are recorded in a database. The data includes the date, time, user response, emotional state, and analysis results. The database is constantly monitored, and an algorithm for detecting anomalies flags consecutive lethargic responses or emotional disturbances as abnormal. If an anomaly is detected, the server automatically notifies the local government via email or SMS to promptly report the user's status.

[0691] Furthermore, the device provides feedback based on the collected conversation and emotional data. For example, if a user is consistently lethargic, the device will check on the user's current situation by saying something like, "You seem a little down lately. Is there anything I can help you with?" and then provide necessary advice and support.

[0692] As a concrete example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if a mental abnormality is detected by the anomaly detection algorithm, a notification is sent to the local government. Subsequently, the local government promptly visits User B's home and takes appropriate action.

[0693] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, achieving more accurate monitoring of health and mental state. As a result, it provides an effective system that improves the quality of life for lonely elderly people and prevents lonely deaths and the progression of dementia.

[0694] The following describes the processing flow.

[0695] Step 1:

[0696] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[0697] Step 2:

[0698] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[0699] Step 3:

[0700] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[0701] Step 4:

[0702] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[0703] Step 5:

[0704] During a conversation, the device analyzes the user's voice and facial expressions using an emotion engine. It identifies the user's emotional state from factors such as voice tone, speaking style, and facial expressions.

[0705] Step 6:

[0706] The device encrypts the recorded dialogue and emotion data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[0707] Step 7:

[0708] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. The server considers both the user's responses and their emotional state when making the evaluation.

[0709] Step 8:

[0710] The server records the analysis results in a database. The stored data includes the date, time, user response, emotional state, and analysis results.

[0711] Step 9:

[0712] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[0713] Step 10:

[0714] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[0715] Step 11:

[0716] The device provides feedback to the user. Based on past conversation data and emotional data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything bothering you?"

[0717] This series of processing steps allows the system, combined with the emotion engine, to monitor the user's health and mental state with greater accuracy and respond quickly when an anomaly is detected.

[0718] (Example 2)

[0719] Next, we will describe Example 2. 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".

[0720] In modern society, health problems caused by loneliness and mental stress are becoming increasingly serious. Loneliness and mental fatigue, especially among the elderly, are significant social issues, requiring rapid and accurate monitoring and appropriate responses. However, conventional systems have limited means of evaluating users' health and mental states, and lack emotional analysis, making it difficult to accurately detect users' conditions. Furthermore, insufficient means of detecting abnormalities often led to delays in prompt responses.

[0721] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for authenticating the user, means for generating and managing conversations with the user using natural language processing, means for evaluating the user's health and mental state by analyzing conversation data, voice data, and image data, means for storing the analysis results in a database and notifying when an abnormality is detected, and means for identifying the user's emotional state using an emotion engine. This makes it possible to monitor the user's health and mental state in detail, including their emotional state, and to take a quick and appropriate response.

[0722] "User authentication" refers to the method of verifying a user's identity when they access a system, using technologies such as cameras or facial recognition.

[0723] "Natural language processing" is a technology that enables computers to understand, analyze, and generate human language, and is used as a means to generate and manage interactions with users.

[0724] "Dialogue data" refers to data that records the content of conversations that take place between a user and a system.

[0725] "Voice data" refers to data used to record and analyze the voice spoken by a user.

[0726] "Image data" refers to data used to record and analyze a user's face and facial expressions.

[0727] "Health status" refers to the user's physical and mental health condition.

[0728] "Mental state" refers to an assessment of the user's mental state and emotional stability.

[0729] "Analysis results" refer to evaluation results regarding the user's health and mental state obtained from dialogue data and emotional data.

[0730] A "database" is a data structure used to store analysis results and interaction data so that they can be referenced and analyzed later.

[0731] "Means for notifying when an anomaly is detected" refers to a function that automatically sends alerts or notifications to relevant organizations and individuals when an anomaly is detected from the analysis results.

[0732] An "emotion engine" is a general term for algorithms and technologies used to identify a user's emotional state from their voice and facial expressions.

[0733] This invention provides a system for routinely monitoring a user's health and mental state and for rapidly detecting abnormalities. This system is implemented using the hardware and software described below.

[0734] Hardware and software:

[0735] 1. Terminal:

[0736] Camera (using facial recognition technology)

[0737] Microphone (voice data collection)

[0738] Display (for interacting with the user)

[0739] 2. Server:

[0740] Database (for storing analysis results and dialogue data)

[0741] Natural language processing models (generative AI models, e.g., GPT-3)

[0742] Emotion engine (face recognition and voice analysis)

[0743] Program processing:

[0744] User authentication:

[0745] The device automatically activates at a designated time each morning and uses its camera to scan the user's face. This authentication process utilizes facial recognition technology (e.g., OpenCV or FaceNet).

[0746] Everyday conversation session:

[0747] Upon successful authentication, the device activates a generative AI model and initiates a casual conversation session with the user using natural language processing technology. Examples of specific questions include, "Did you sleep well last night?"

[0748] Emotion analysis:

[0749] During user responses, the emotion engine analyzes voice tone and facial expressions in real time. For example, technologies such as Deep Learning for Audio are used to analyze emotions from audio data, and OpenCV and deep learning models are used to analyze emotions from facial expressions.

[0750] Data encryption and transmission:

[0751] Emotional and conversational data are converted into text format and encrypted using encryption technologies such as AES. The encrypted data is then sent to the server.

[0752] Server-based analysis:

[0753] The server receives encrypted data and performs decryption. It uses natural language processing and an emotion engine to evaluate the user's health and mental state. Specifically, both the response content and emotional state are considered when evaluating health.

[0754] Recording to the database:

[0755] The analysis results are stored in a database, recording the date, time, response content, emotional state, and analysis results. The database is constantly monitored, and algorithms for detecting anomalies monitor for consecutive lethargic responses and emotional disturbances.

[0756] Anomaly detection and notification:

[0757] If an anomaly is detected, the server will automatically notify the local government via email or SMS. Specifically, notifications will be sent using the Twilio API, among other methods.

[0758] Provide feedback:

[0759] Based on the collected data, the device's AI model generates and provides necessary feedback to the user. This feedback includes conversational phrases such as, "You seem a little down lately. Is there anything I can help you with?"

[0760] Specific example:

[0761] For example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if the anomaly detection algorithm detects a mental abnormality, a notification is sent to the local government. The local government then promptly visits User B's home and takes appropriate action.

[0762] Examples of prompts for a generative AI model:

[0763] Enter a question such as, "Did you sleep well last night?"

[0764] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, enabling more accurate monitoring of health and mental state.

[0765] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0766] Step 1:

[0767] Device startup and user authentication

[0768] Input: Time (specified date and time)

[0769] Operation: The device will automatically start up and turn on the camera at a specified time every morning.

[0770] Data processing: The user's face is scanned using facial recognition technology (e.g., OpenCV or FaceNet) and compared with registered user data.

[0771] Output: User authentication result (success / failure)

[0772] Step 2:

[0773] Starting a daily conversation session

[0774] Input: User authentication result (success)

[0775] Operation: Upon successful authentication, the device launches a generated AI model (e.g., GPT-3) and begins a casual conversation with the user using natural language processing technology.

[0776] Prompt: Good morning. Did you sleep well last night?

[0777] Output: User response (text format)

[0778] Step 3:

[0779] User response collection and sentiment analysis

[0780] Input: User response (voice, facial expression)

[0781] Operation: The emotion engine analyzes the user's voice tone and facial expressions in real time. Deep Learning for Audio is used for audio data, and OpenCV and deep learning models are used for facial expression data.

[0782] Data processing: Convert emotional data (tone, facial expression) to text format.

[0783] Output: Sentiment data (text format)

[0784] Step 4:

[0785] Data encryption and transmission

[0786] Input: Dialogue data, emotion data (text format)

[0787] Operation: Encrypts emotional data and dialogue data using encryption technologies such as AES.

[0788] Data processing: Generation of encrypted data

[0789] Output: Encrypted data

[0790] Step 5:

[0791] Decryption of data by the server

[0792] Input: Encrypted data

[0793] Operation: The server receives the transmitted encrypted data and decrypts it using a decryption algorithm such as AES.

[0794] Data processing: Generation of decoded data

[0795] Output: Decoded data (dialogue data, sentiment data)

[0796] Step 6:

[0797] Assessment of physical and mental health

[0798] Input: Dialogue data, emotion data (decoded data)

[0799] Operation: The server uses natural language processing and an emotion engine to evaluate the user's health and mental state. Specifically, it analyzes the response using NLP technology and combines it with emotion data to perform a comprehensive evaluation.

[0800] Data processing: Generation of health status assessments

[0801] Output: Assessment results of health and mental state

[0802] Step 7:

[0803] Database recording and anomaly detection

[0804] Input: Assessment results of health and mental state

[0805] Operation: The analysis results are saved to a database. The database is constantly monitored, and an algorithm detects anomalies, monitoring for continuous lethargic responses and emotional disturbances.

[0806] Data processing: Execution of anomaly detection algorithms

[0807] Output: Anomaly detection result (normal / abnormal)

[0808] Step 8:

[0809] Notification to local government

[0810] Input: Anomaly detection result (anomaly)

[0811] Operation: If an anomaly is detected, the server will use the Twilio API or similar tools to notify the local government via email or SMS.

[0812] Data processing: Generating notification messages

[0813] Output: Notification to local government

[0814] Step 9:

[0815] Provide feedback

[0816] Input: Dialogue data, emotion data (historical data)

[0817] Operation: Based on the collected data, the device uses a generated AI model to provide necessary feedback to the user.

[0818] Prompt: You seem a little down lately. Is there anything I can help you with?

[0819] Output: Feedback content (text format)

[0820] (Application Example 2)

[0821] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0822] Conventional health and mental state monitoring systems have struggled to grasp users' emotional and health states in real time during their daily activities and to take appropriate action quickly based on that information. Furthermore, in public spaces such as stores, there have been challenges in appropriately understanding customers' emotional states and providing services based on that understanding. The present invention aims to solve these problems and provide a system that can evaluate the health and mental state of users and customers in real time and enable a rapid response when an abnormality is detected.

[0823] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0824] In this invention, the server includes means for authenticating users, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for monitoring the emotional and health states of customers in real time within the store, means for notifying store staff of the results of the customer emotion analysis and health evaluation, and means for providing customer service feedback according to the customer's state. This enables real-time monitoring of the emotional and health states of users and customers, and appropriate responses and feedback based on these.

[0825] Definitions of important words

[0826] "User authentication" is a means by which a system verifies the user's identity.

[0827] Natural language processing is a technology that enables computers to understand, generate, and manage human language.

[0828] "Dialogue data" refers to digital data that includes the content of conversations between users and systems.

[0829] "Health status" refers to information about the user's physical condition and health.

[0830] "Mental state" refers to the user's psychological or emotional state.

[0831] A "database" is a storage system for systematically saving analysis results and data.

[0832] Anomaly detection is the process of detecting unusual states or behaviors.

[0833] "Notification" refers to the act of a system informing users or administrators of specific information.

[0834] "Inside the store" refers to the physical commercial space.

[0835] "Customer" refers to an individual who uses a store or service.

[0836] Real-time monitoring is the process of collecting and analyzing information and data without delay.

[0837] "Emotional analysis" is a technology that evaluates an individual's emotional state through the analysis of voice and video.

[0838] "Health assessment results" refer to the analysis results regarding the health status of users and customers.

[0839] "Feedback" is the act of a system providing responses or advice to a user.

[0840] "Customer service" refers to the process of providing product descriptions, assistance, and customer service to customers.

[0841] Modes for carrying out the invention

[0842] This invention provides a system for routinely monitoring the health and mental state of users and customers and for rapidly detecting abnormalities. A key feature is its integration with an emotion engine, which allows for the evaluation of the emotional state of users and customers. Specific embodiments of this system are described below.

[0843] This system is used in commercial spaces (such as stores) where users and customers are present, allowing store employees to monitor customers' emotional and health states in real time using smart glasses. The main components of the system are as follows:

[0844] 1. User Authentication Method: The system uses facial recognition technology to verify the identity of users and customers. Facial recognition software will utilize libraries such as OpenCV.

[0845] 2. Natural Language Processing (NLP) Methods: The system uses NLP libraries such as Transformers and the BERT model to generate and manage conversations with users and customers. This allows for the analysis of conversation content and the evaluation of health and mental state.

[0846] 3. Emotion Analysis Methods: An emotion engine is used to analyze audio and video data and evaluate the customer's emotional state. This includes facial expression analysis and voice analysis.

[0847] 4. Database: A storage system is used to save the analyzed results. Data on customers' emotional and health states is accumulated, and an anomaly detection algorithm is applied if an anomaly is detected.

[0848] 5. Notification method: If an anomaly is detected, the server will send a notification to the store staff using a communication service such as Twilio.

[0849] 6. Real-time monitoring method: Smart glasses monitor the customer's emotional state and health status in real time and provide feedback to the store staff.

[0850] By combining the above components, it is possible to monitor the health and mental state of users and customers on a daily basis and to quickly notify them if any abnormalities are detected.

[0851] Usage example

[0852] As a concrete example, consider a scenario where a store monitors the health and emotional state of its customers. The store staff wear smart glasses. When a customer enters the store to look at merchandise, facial recognition technology identifies the customer's face, and a dialogue using natural language processing technology begins. The smart glasses analyze the customer's emotional state in real time, and data indicating that "the customer is experiencing stress" is obtained. If appropriate customer service is needed, an alert is sent to the staff via Twilio, allowing them to respond quickly and appropriately.

[0853] Example of a prompt

[0854] "Imagine a customer facial recognition system for emotion analysis. Develop an application that detects customer stress and distress, and supports appropriate feedback and responses."

[0855] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0856] Program processing steps

[0857] Step 1:

[0858] The smart glasses activate the camera and capture images of customers inside the store.

[0859] (Specific action)

[0860] The smart glasses' camera captures video in real time, and that video data is sent to the system.

[0861] (Input) Customer's real-time video data

[0862] (Output) Face feature data required for face recognition

[0863] Step 2:

[0864] The server uses facial recognition technology to identify the customer's face.

[0865] (Specific action)

[0866] The server uses the OpenCV library to perform face recognition on video data and extract customer facial features.

[0867] (Input) Facial feature data

[0868] (Output) Recognized face location information and features

[0869] Step 3:

[0870] The server activates a natural language processing model and generates and manages interactions with the user.

[0871] (Specific action)

[0872] The server uses the Transformers library and the BERT model to analyze conversations between store employees and customers in real time.

[0873] (Input) Text data of conversations with customers

[0874] (Output) Analyzed emotional and health assessment results

[0875] Step 4:

[0876] The server uses an emotion engine to analyze the customer's emotional state.

[0877] (Specific action)

[0878] The server analyzes audio and video data to assess the customer's emotional state in real time. This includes voice tone analysis and facial expression analysis.

[0879] (Input) Voice data, facial expression data

[0880] (Output) Customer's emotional state

[0881] Step 5:

[0882] The server saves the emotion analysis results to a database and detects anomalies.

[0883] (Specific action)

[0884] The server records the emotion analysis results in a database and applies an algorithm to detect consecutive abnormal emotional states.

[0885] (Input) Sentiment analysis result data

[0886] (Output) Anomaly flag and its cause data

[0887] Step 6:

[0888] The server will send a notification based on the anomaly detection results.

[0889] (Specific action)

[0890] The server uses Twilio to send notifications to store employees informing them of unusual customer behavior.

[0891] (Input) Anomaly detection flag

[0892] (Output) Notification message to store staff

[0893] Step 7:

[0894] Smart glasses provide feedback.

[0895] (Specific action)

[0896] The smart glasses display information about the customer's emotional state and hints for how to respond, providing feedback to the store staff.

[0897] (Input) Customer sentiment assessment data, health assessment data

[0898] (Output) Feedback content

[0899] Step 8:

[0900] Store staff will provide appropriate service to customers within the store.

[0901] (Specific action)

[0902] Store staff provide appropriate service and support to customers based on feedback from smart glasses.

[0903] (Input) Feedback content

[0904] (Output) Customer service and support

[0905] The above outlines the specific processing steps for implementing the invention.

[0906] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0907] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0908] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0909] [Third Embodiment]

[0910] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0911] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0912] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0913] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0914] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0915] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0916] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0917] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0918] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0919] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0920] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0921] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0922] This invention provides a system that offers daily support to lonely elderly people and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[0923] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. It also includes means for encrypting interaction data and sending it to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, and means for providing feedback based on the interaction logs.

[0924] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[0925] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" The user's responses are recorded by the device and converted into text. This text data is then encrypted and sent to the server.

[0926] The server analyzes the received data and uses natural language processing to extract keywords and perform sentiment analysis. Based on the analysis results, it evaluates the user's health and mental state. The analysis results are recorded in a database.

[0927] The results stored in the database are constantly monitored by an algorithm designed to detect anomalies. If an anomaly is detected, the server automatically sends a notification to the local government, issuing a warning via email or SMS. This notification includes the username, the date and time the problem occurred, and a brief analysis.

[0928] Furthermore, the device provides feedback to the user based on the collected conversational data. For example, the device generates feedback to reconfirm the user's mental state, such as, "You seem a little down lately. Is there anything bothering you?"

[0929] As a concrete example, consider a scenario where User A interacts with the device daily. User A replies, "Yes, I slept well," and then the device asks, "What are your plans for today?" User A replies, "I don't have any particular plans for today." These responses are sent to the server, and the system evaluates User A's health as good. However, if User A shows lethargic responses a few days later, the server detects an abnormality in their mental state and sends a notification to the local government. In this way, abnormalities can be detected early, enabling appropriate intervention.

[0930] As described above, the present invention provides an effective system for preventing lonely deaths and the progression of dementia by routinely monitoring the health and mental state of lonely elderly people and detecting abnormalities early.

[0931] The following describes the processing flow.

[0932] Step 1:

[0933] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[0934] Step 2:

[0935] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[0936] Step 3:

[0937] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[0938] Step 4:

[0939] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[0940] Step 5:

[0941] The device encrypts the recorded text-based conversation data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[0942] Step 6:

[0943] The server decrypts the received encrypted data and uses natural language processing (NLP) techniques to extract keywords and perform sentiment analysis. Based on the user's responses, the server evaluates their health and mental state.

[0944] Step 7:

[0945] The server records the analysis results in a database. The stored data includes the date, time, user response, and analysis results.

[0946] Step 8:

[0947] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[0948] Step 9:

[0949] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[0950] Step 10:

[0951] The device provides feedback to the user. Based on past conversation data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything I can help you with?"

[0952] This processing step allows for effective monitoring of the health and safety of elderly people living alone, and enables a rapid response if any abnormalities are detected.

[0953] (Example 1)

[0954] Next, we will describe Example 1. 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."

[0955] In modern society, it is crucial to detect abnormalities in the health and mental state of lonely elderly individuals early and provide appropriate support. However, conventional systems rely on elderly individuals reporting abnormalities themselves, making early detection difficult. Furthermore, the collection and analysis of conversational data is insufficient, making it difficult to accurately monitor changes in individual health and mental states.

[0956] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0957] In this invention, the server includes means for the terminal to automatically start up at a specified time each morning and authenticate the user by scanning their face; means for generating and managing conversations with the user using natural language processing; means for recording the user's voice responses, converting them to text format and encrypting them; means for transmitting the encrypted text data to the server; means for the server to analyze the received data and perform keyword extraction and sentiment analysis using natural language processing; means for evaluating the user's health and mental state based on the analysis results and recording it in a database; and means for continuously monitoring using an algorithm for detecting abnormalities and notifying the local government if an abnormality is detected. This makes it possible to monitor the health and mental state of lonely elderly people on a daily basis, detect abnormalities early, and take appropriate action.

[0958] A "terminal" is an electronic device that functions as the user interface of a system and manages interactions with the user.

[0959] "User authentication" is a process of verifying a user's identity and is a security function that uses technologies such as facial recognition.

[0960] Natural language processing is a technology that enables computers to understand, interpret, and generate human language, and is used to generate and manage interactions with users.

[0961] "Speech recognition" is a technology that converts speech into text data, and is a technology for processing a user's verbal responses as written information.

[0962] "Encryption" is the process of hiding the contents of data in order to transfer it securely, and it is carried out using a specific algorithm.

[0963] A "server" is a centralized computer system that analyzes received data and stores and manages the results.

[0964] "Data analysis" is the process of processing collected data to derive useful information and insights, and it involves techniques such as natural language processing and other analytical methods.

[0965] "Sentiment analysis" is a technology that identifies and classifies emotions from text and audio data, and is used to evaluate a user's mental state.

[0966] A "database" is a system for efficiently storing, searching, and managing data, and is used to record analytical results.

[0967] Anomaly detection is a technique for identifying data that deviates from normal patterns, and is a process for determining whether there is an abnormality in one's health or mental state.

[0968] "Feedback" refers to advice and confirmations that a system provides based on its interaction with the user and analysis results, and includes messages that include reactions and suggestions to the user.

[0969] "Local government notification" is a function that contacts local government agencies when an anomaly is detected, and this is done via email or SMS.

[0970] Modes for carrying out the invention

[0971] This invention provides a system that offers daily support to lonely elderly individuals and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[0972] System Configuration

[0973] The system consists mainly of the following elements:

[0974] 1. Terminal

[0975] The terminal is an electronic device installed in the homes of elderly people to manage interactions with the user.

[0976] It automatically starts up at a designated time every morning and uses facial recognition technology to authenticate the user.

[0977] Conversations with users are conducted using a natural language processing engine (e.g., OpenAI GPT-3).

[0978] Use speech recognition software (e.g., Google Cloud Speech-to-Text) to convert the user's voice response into text.

[0979] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption) and sent to the server using a secure protocol (e.g., HTTPS).

[0980] 2. Server

[0981] A server is a centralized computer system that analyzes received data and stores and manages the results.

[0982] We use natural language processing engines (e.g., NLTK and spaCy) to extract keywords and perform sentiment analysis to evaluate the user's health and mental state.

[0983] The analysis results are recorded in an SQL database (e.g., MySQL or PostgreSQL) and continuously monitored by anomaly detection algorithms (e.g., machine learning models).

[0984] If an anomaly is detected, the system will notify the local government via email or SMS API (e.g., Twilio).

[0985] Specific operation of the system

[0986] 1. User Authentication

[0987] The device automatically starts up at 8 AM every morning and scans the user's face with its built-in camera. It also authenticates the user using a facial recognition algorithm (e.g., OpenCV or a common facial recognition library).

[0988] 2. Everyday Conversation Session

[0989] Upon successful authentication, the device activates a generative AI and begins a casual conversation with the user.

[0990] For example, questions such as "Did you sleep well last night?" and "What are your plans for today?" are generated.

[0991] 3. Data collection, encryption, and transmission

[0992] The device records the user's voice response and converts it into text data using speech recognition software.

[0993] The converted text data is encrypted and sent to the server using a secure protocol.

[0994] 4. Data Analysis and Anomaly Detection

[0995] The server analyzes the received data and uses natural language processing to evaluate the user's health and mental state.

[0996] The analysis results are recorded in a database and continuously monitored using an anomaly detection algorithm.

[0997] If an anomaly is detected, a notification will be sent to the local government via email or SMS.

[0998] 5. Provide feedback

[0999] The device provides appropriate feedback to the user based on the analysis results. It uses generative AI to generate the feedback content and converts it into speech using speech synthesis software (e.g., Amazon Polly or Google Text-to-Speech).

[1000] For example, feedback such as, "You seem a little down lately. Is there anything I can help you with?" might be provided.

[1001] Specific example

[1002] Example of interaction with User A:

[1003] Every morning at 8:00 AM, the terminal automatically starts up and scans user A's face to perform authentication.

[1004] Upon successful authentication, the device asks, "Did you sleep well last night?", and User A replies, "Yes, I slept very well."

[1005] This response will be converted to text, encrypted, and sent to the server.

[1006] The server analyzes the data and determines that User A's health status is good.

[1007] A few days later, if user A responds with "I'm tired and don't want to do anything," the server will use that data to detect an anomaly and send a notification to the local government.

[1008] The device provides feedback saying, "You seem a little down lately. Is there anything I can help you with?"

[1009] Thus, the present invention provides a system for routinely monitoring the health and mental state of lonely elderly people, detecting abnormalities early, and providing appropriate intervention.

[1010] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1011] Step 1:

[1012] System startup and user authentication

[1013] Input: System startup time, user's face image

[1014] Output: Authentication success / failure result

[1015] The device is set to automatically start up at a specified time every morning (e.g., 8:00 AM).

[1016] After startup, the built-in camera is used to capture an image of the user's face.

[1017] Using facial recognition technology (e.g., OpenCV or a general facial recognition library), this facial image is compared against pre-registered facial data.

[1018] If authentication is successful, a "Authentication successful" message is sent to the system, and the process proceeds to the next step. If it fails, the system will either retry or issue an alert.

[1019] Step 2:

[1020] Starting a daily conversation session

[1021] Input: Authentication success message

[1022] Output: Generated question prompt, user voice response

[1023] After successful authentication, the terminal launches a generated AI model (e.g., OpenAI GPT-3).

[1024] To begin speaking, generate a prompt sentence (e.g., "Did you sleep well last night?").

[1025] This question is output to the user as an audio message, and the user's voice response is collected.

[1026] Step 3:

[1027] Speech-to-text conversion and encryption

[1028] Input: User voice response

[1029] Output: Encrypted text data

[1030] The device records the collected voice responses using its microphone.

[1031] Use speech recognition software (e.g., Google Cloud Speech-to-Text) to convert recorded audio into text data.

[1032] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption).

[1033] Create encrypted text data and prepare to proceed to the next step.

[1034] Step 4:

[1035] Sending data to the server

[1036] Input: Encrypted text data

[1037] Output: Encrypted data sent to the server

[1038] The device sends encrypted text data to the server using a secure protocol (e.g., HTTPS).

[1039] The server checks the received data and prepares for the next analysis step.

[1040] Step 5:

[1041] Data Analysis

[1042] Input: Encrypted data sent to the server

[1043] Output: Analysis results (assessment of health and mental state)

[1044] The server decrypts the received encrypted data to obtain the text data.

[1045] We will analyze text data using a natural language processing engine (e.g., NLTK or spaCy). Specifically, we will perform keyword extraction and sentiment analysis.

[1046] Based on the analysis results, data is generated to evaluate the user's health and mental state. The evaluation results are then used in the next step.

[1047] Step 6:

[1048] Database recording of analysis results and anomaly detection.

[1049] Input: Analysis results

[1050] Output: Analysis results recorded in the database, anomaly detection notifications.

[1051] The server records the analysis results in an SQL database (e.g., MySQL or PostgreSQL).

[1052] Anomaly detection algorithms (e.g., machine learning models) are used to continuously monitor the analysis results stored in the database.

[1053] If an anomaly is detected, the notification system is automatically activated and sends notifications to relevant parties via email or SMS API (e.g., Twilio).

[1054] Step 7:

[1055] Provide feedback

[1056] Input: Analysis results and evaluation of anomaly detection

[1057] Output: Feedback provided to the user

[1058] The device uses a generative AI model to generate appropriate feedback (e.g., "You seem a little down lately. Is there anything I can help you with?").

[1059] Feedback generated using text-to-speech software (e.g., Amazon Polly or Google Text-to-Speech) is converted into speech and provided to the user.

[1060] The above describes the specific processing flow of this system's program.

[1061] (Application Example 1)

[1062] Next, we will explain Application Example 1. In the following explanation, 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."

[1063] In modern society, monitoring the loneliness and health status of the elderly is a critical issue. In particular, the number of elderly people who are unable to receive prompt and appropriate support is increasing. Such situations can lead to serious problems such as deteriorating health and lonely deaths, necessitating effective countermeasures. Furthermore, there is a need for systems used daily by the elderly that efficiently monitor their health and mental state and respond quickly when abnormalities are detected.

[1064] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1065] In this invention, the server includes means for authenticating the user, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for performing user authentication and health checks using facial recognition technology and voice interaction, means for performing sentiment analysis of user responses using natural language processing, and means for automatically detecting anomalies and providing notifications based on the sentiment analysis results. This makes it possible to routinely monitor the health and mental state of elderly people and to quickly notify them if an anomaly is detected.

[1066] "Means of user authentication" refers to a system that uses cameras or biometric authentication technology to identify users and verify their identity.

[1067] "Means for generating and managing user interactions using natural language processing" refers to technologies that enable smooth conversations by allowing computers to understand natural language and generate and manage responses.

[1068] "Methods for analyzing dialogue data to evaluate a user's health and mental state" refers to algorithms that analyze dialogue content to diagnose and evaluate a user's health and mental state.

[1069] "A means of saving analysis results to a database and notifying when an anomaly is detected" refers to a system that stores the results of data analysis in a database and notifies when an anomaly is discovered.

[1070] "A means of performing user authentication and health checks using facial recognition technology and voice dialogue" refers to a system that utilizes facial recognition and voice dialogue technology to authenticate users and verify their health status.

[1071] "Means for performing sentiment analysis of user responses using natural language processing" refers to a technology that uses natural language processing techniques to analyze user responses and evaluate their emotional state.

[1072] "A means of automatically detecting anomalies and providing notifications based on sentiment analysis results" refers to a system that uses the results of sentiment analysis to detect anomalies and automatically provides notifications as needed.

[1073] "Regularly monitoring the user's health and mental state and promptly notifying them if an abnormality is detected" means constantly monitoring the user's health and mental state and immediately notifying relevant parties when an abnormality is discovered.

[1074] This invention is a system that routinely monitors the health and mental state of elderly individuals and promptly notifies them if any abnormalities are detected. Specific embodiments of this system are described below.

[1075] 1. System Configuration

[1076] The system consists of the following main elements:

[1077] User authentication method (facial recognition technology)

[1078] A means of generating and managing user interactions using natural language processing.

[1079] A method for analyzing dialogue data to evaluate the user's health and mental state.

[1080] A means of saving analysis results to a database and notifying when an anomaly is detected.

[1081] A method for user authentication and health checks using facial recognition technology and voice interaction.

[1082] A method for performing sentiment analysis of user responses using natural language processing.

[1083] A method for automatically detecting anomalies and sending notifications based on sentiment analysis results.

[1084] 2. Use of Hardware and Software

[1085] hardware

[1086] Camera (OpenCV compatible)

[1087] Microphone (compatible with speech_recognition library)

[1088] software

[1089] OpenCV: Uses facial recognition technology to authenticate users.

[1090] speech_recognition: Enables voice interaction and checks the user's health status.

[1091] nltk: Uses natural language processing to analyze user response text and perform sentiment analysis.

[1092] requests: Send analysis results to the remote server.

[1093] 3. Data Processing and Analysis

[1094] Dialogue data generated within the system is converted into text using speech recognition technology and further analyzed using natural language processing. During this process, sentiment analysis is performed to assess the user's health and mental state. This evaluation data is stored in a database, and if an anomaly is detected, the system automatically notifies the local government and family.

[1095] 4. Specific Examples

[1096] Example 1: User authentication and health check

[1097] The user is authenticated by facial recognition technology when they stand in front of the camera. After authentication, the system asks aloud, "Did you sleep well last night?", to which the user replies, "Yes, I slept well." This response is converted into text by speech recognition technology and sent to a server for analysis.

[1098] Example 2: Sentiment analysis and notification

[1099] If a user responds with "I don't feel like doing anything in particular today," a negative score is obtained through sentiment analysis using natural language processing. Based on this analysis, the server detects an anomaly and sends notifications to local authorities and family members.

[1100] Examples of prompts for generative AI models

[1101] Use the following text to analyze the user's health status and calculate their emotional score:

[1102] "I slept well last night."

[1103] In this way, the system can routinely monitor the health and mental state of elderly individuals and respond quickly in the event of an emergency. This enables effective interventions to prevent lonely deaths and the progression of dementia.

[1104] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1105] Step 1:

[1106] The device automatically starts up at a designated time each morning and scans the user's face using its camera. The input is the camera image, and the output is the result of user authentication. User authentication is performed using facial recognition technology (OpenCV).

[1107] Step 2:

[1108] If user authentication is successful, the device will prompt for voice input and begin checking the user's health status. It will ask questions such as, "Did you sleep well last night?" The input for this step is the user's voice response, and the output is audio data.

[1109] Step 3:

[1110] The device uses speech recognition technology (speech_recognition library) to convert the user's voice responses into text data. The input is voice data, and the output is text data.

[1111] Step 4:

[1112] Text data is sent from the terminal to the server in an encrypted state. The input here is text data, and the output is encrypted data. The requests library is used for sending.

[1113] Step 5:

[1114] The server analyzes the received text data using natural language processing (NLP) techniques (the nltk library). The input is encrypted text data, and the output is the sentiment analysis result. Specifically, it calculates a sentiment score from the text data.

[1115] Step 6:

[1116] The server evaluates the user's health and mental state based on the analysis results and saves the results to a database. The input is the emotion analysis results, and the output is the operation to save them to the database.

[1117] Step 7:

[1118] The server periodically monitors the database and sends notifications to local governments and families if an anomaly is detected. The input for this step is the analysis results from the database, and the output is a notification email or SMS.

[1119] Step 8:

[1120] The device provides feedback to the user. It generates feedback in the form of, "You seem a little down lately. Is there anything I can help you with?" The input for this step is the result of sentiment analysis, and the output is voice feedback to the user.

[1121] Step 9:

[1122] The prompt text used as input to the generative AI model is: "Analyze the user's health status using the following text and calculate a sentiment score: 'I slept well last night.'" The input is the prompt text, and the output is the generated sentiment score.

[1123] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1124] This invention relates to a system for routinely monitoring a user's health and mental state and rapidly detecting abnormalities. It is particularly characterized by its integration with an emotion engine, which allows the system to also evaluate the user's emotional state. Specific embodiments of this system are described below.

[1125] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. In addition, it includes means for encrypting and sending interaction data to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, means for providing feedback based on the interaction logs, and an emotion engine that recognizes the user's emotions.

[1126] The emotion engine analyzes audio and image data collected during interactions with the user to identify the user's emotional state. Specifically, the emotion engine uses algorithms that estimate emotions from the user's voice tone, speaking style, facial expressions, and other factors.

[1127] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[1128] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" In addition, an emotion engine analyzes the user's emotions in real time from their responses, analyzing the tone of voice and facial expressions when the user answers, "Yes, I slept well." The analysis results are converted into text format as emotion data, encrypted along with the dialogue data, and then sent to the server.

[1129] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. Specifically, both the user's responses and emotional state are considered in the health assessment. For example, even if a user answers "I slept well," if their tone of voice or facial expression is listless, they will be evaluated as mentally exhausted.

[1130] The analysis results are recorded in a database. The data includes the date, time, user response, emotional state, and analysis results. The database is constantly monitored, and an algorithm for detecting anomalies flags consecutive lethargic responses or emotional disturbances as abnormal. If an anomaly is detected, the server automatically notifies the local government via email or SMS to promptly report the user's status.

[1131] Furthermore, the device provides feedback based on the collected conversation and emotional data. For example, if a user is consistently lethargic, the device will check on the user's current situation by saying something like, "You seem a little down lately. Is there anything I can help you with?" and then provide necessary advice and support.

[1132] As a concrete example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if a mental abnormality is detected by the anomaly detection algorithm, a notification is sent to the local government. Subsequently, the local government promptly visits User B's home and takes appropriate action.

[1133] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, achieving more accurate monitoring of health and mental state. As a result, it provides an effective system that improves the quality of life for lonely elderly people and prevents lonely deaths and the progression of dementia.

[1134] The following describes the processing flow.

[1135] Step 1:

[1136] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[1137] Step 2:

[1138] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[1139] Step 3:

[1140] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[1141] Step 4:

[1142] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[1143] Step 5:

[1144] During a conversation, the device analyzes the user's voice and facial expressions using an emotion engine. It identifies the user's emotional state from factors such as voice tone, speaking style, and facial expressions.

[1145] Step 6:

[1146] The device encrypts the recorded dialogue and emotion data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[1147] Step 7:

[1148] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. The server considers both the user's responses and their emotional state when making the evaluation.

[1149] Step 8:

[1150] The server records the analysis results in a database. The stored data includes the date, time, user response, emotional state, and analysis results.

[1151] Step 9:

[1152] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[1153] Step 10:

[1154] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[1155] Step 11:

[1156] The device provides feedback to the user. Based on past conversation data and emotional data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything bothering you?"

[1157] This series of processing steps allows the system, combined with the emotion engine, to monitor the user's health and mental state with greater accuracy and respond quickly when an anomaly is detected.

[1158] (Example 2)

[1159] Next, we will describe Example 2. 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."

[1160] In modern society, health problems caused by loneliness and mental stress are becoming increasingly serious. Loneliness and mental fatigue, especially among the elderly, are significant social issues, requiring rapid and accurate monitoring and appropriate responses. However, conventional systems have limited means of evaluating users' health and mental states, and lack emotional analysis, making it difficult to accurately detect users' conditions. Furthermore, insufficient means of detecting abnormalities often led to delays in prompt responses.

[1161] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for authenticating the user, means for generating and managing conversations with the user using natural language processing, means for evaluating the user's health and mental state by analyzing conversation data, voice data, and image data, means for storing the analysis results in a database and notifying when an abnormality is detected, and means for identifying the user's emotional state using an emotion engine. This makes it possible to monitor the user's health and mental state in detail, including their emotional state, and to take a quick and appropriate response.

[1162] "User authentication" refers to the method of verifying a user's identity when they access a system, using technologies such as cameras or facial recognition.

[1163] "Natural language processing" is a technology that enables computers to understand, analyze, and generate human language, and is used as a means to generate and manage interactions with users.

[1164] "Dialogue data" refers to data that records the content of conversations that take place between a user and a system.

[1165] "Voice data" refers to data used to record and analyze the voice spoken by a user.

[1166] "Image data" refers to data used to record and analyze a user's face and facial expressions.

[1167] "Health status" refers to the user's physical and mental health condition.

[1168] "Mental state" refers to an assessment of the user's mental state and emotional stability.

[1169] "Analysis results" refer to evaluation results regarding the user's health and mental state obtained from dialogue data and emotional data.

[1170] A "database" is a data structure used to store analysis results and interaction data so that they can be referenced and analyzed later.

[1171] "Means for notifying when an anomaly is detected" refers to a function that automatically sends alerts or notifications to relevant organizations and individuals when an anomaly is detected from the analysis results.

[1172] An "emotion engine" is a general term for algorithms and technologies used to identify a user's emotional state from their voice and facial expressions.

[1173] This invention provides a system for routinely monitoring a user's health and mental state and for rapidly detecting abnormalities. This system is implemented using the hardware and software described below.

[1174] Hardware and software:

[1175] 1. Terminal:

[1176] Camera (using facial recognition technology)

[1177] Microphone (voice data collection)

[1178] Display (for interacting with the user)

[1179] 2. Server:

[1180] Database (for storing analysis results and dialogue data)

[1181] Natural language processing models (generative AI models, e.g., GPT-3)

[1182] Emotion engine (face recognition and voice analysis)

[1183] Program processing:

[1184] User authentication:

[1185] The device automatically activates at a designated time each morning and uses its camera to scan the user's face. This authentication process utilizes facial recognition technology (e.g., OpenCV or FaceNet).

[1186] Everyday conversation session:

[1187] Upon successful authentication, the device activates a generative AI model and initiates a casual conversation session with the user using natural language processing technology. Examples of specific questions include, "Did you sleep well last night?"

[1188] Emotion analysis:

[1189] During user responses, the emotion engine analyzes voice tone and facial expressions in real time. For example, technologies such as Deep Learning for Audio are used to analyze emotions from audio data, and OpenCV and deep learning models are used to analyze emotions from facial expressions.

[1190] Data encryption and transmission:

[1191] Emotional and conversational data are converted into text format and encrypted using encryption technologies such as AES. The encrypted data is then sent to the server.

[1192] Server-based analysis:

[1193] The server receives encrypted data and performs decryption. It uses natural language processing and an emotion engine to evaluate the user's health and mental state. Specifically, both the response content and emotional state are considered when evaluating health.

[1194] Recording to the database:

[1195] The analysis results are stored in a database, recording the date, time, response content, emotional state, and analysis results. The database is constantly monitored, and algorithms for detecting anomalies monitor for consecutive lethargic responses and emotional disturbances.

[1196] Anomaly detection and notification:

[1197] If an anomaly is detected, the server will automatically notify the local government via email or SMS. Specifically, notifications will be sent using the Twilio API, among other methods.

[1198] Provide feedback:

[1199] Based on the collected data, the device's AI model generates and provides necessary feedback to the user. This feedback includes conversational phrases such as, "You seem a little down lately. Is there anything I can help you with?"

[1200] Specific example:

[1201] For example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if the anomaly detection algorithm detects a mental abnormality, a notification is sent to the local government. The local government then promptly visits User B's home and takes appropriate action.

[1202] Examples of prompts for a generative AI model:

[1203] Enter a question such as, "Did you sleep well last night?"

[1204] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, enabling more accurate monitoring of health and mental state.

[1205] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1206] Step 1:

[1207] Device startup and user authentication

[1208] Input: Time (specified date and time)

[1209] Operation: The device will automatically start up and turn on the camera at a specified time every morning.

[1210] Data processing: The user's face is scanned using facial recognition technology (e.g., OpenCV or FaceNet) and compared with registered user data.

[1211] Output: User authentication result (success / failure)

[1212] Step 2:

[1213] Starting a daily conversation session

[1214] Input: User authentication result (success)

[1215] Operation: Upon successful authentication, the device launches a generated AI model (e.g., GPT-3) and begins a casual conversation with the user using natural language processing technology.

[1216] Prompt: Good morning. Did you sleep well last night?

[1217] Output: User response (text format)

[1218] Step 3:

[1219] User response collection and sentiment analysis

[1220] Input: User response (voice, facial expression)

[1221] Operation: The emotion engine analyzes the user's voice tone and facial expressions in real time. Deep Learning for Audio is used for audio data, and OpenCV and deep learning models are used for facial expression data.

[1222] Data processing: Convert emotional data (tone, facial expression) to text format.

[1223] Output: Sentiment data (text format)

[1224] Step 4:

[1225] Data encryption and transmission

[1226] Input: Dialogue data, emotion data (text format)

[1227] Operation: Encrypts emotional data and dialogue data using encryption technologies such as AES.

[1228] Data processing: Generation of encrypted data

[1229] Output: Encrypted data

[1230] Step 5:

[1231] Decryption of data by the server

[1232] Input: Encrypted data

[1233] Operation: The server receives the transmitted encrypted data and decrypts it using a decryption algorithm such as AES.

[1234] Data processing: Generation of decoded data

[1235] Output: Decoded data (dialogue data, sentiment data)

[1236] Step 6:

[1237] Assessment of physical and mental health

[1238] Input: Dialogue data, emotion data (decoded data)

[1239] Operation: The server uses natural language processing and an emotion engine to evaluate the user's health and mental state. Specifically, it analyzes the response using NLP technology and combines it with emotion data to perform a comprehensive evaluation.

[1240] Data processing: Generation of health status assessments

[1241] Output: Assessment results of health and mental state

[1242] Step 7:

[1243] Database recording and anomaly detection

[1244] Input: Assessment results of health and mental state

[1245] Operation: The analysis results are saved to a database. The database is constantly monitored, and an algorithm detects anomalies, monitoring for continuous lethargic responses and emotional disturbances.

[1246] Data processing: Execution of anomaly detection algorithms

[1247] Output: Anomaly detection result (normal / abnormal)

[1248] Step 8:

[1249] Notification to local government

[1250] Input: Anomaly detection result (anomaly)

[1251] Operation: If an anomaly is detected, the server will use the Twilio API or similar tools to notify the local government via email or SMS.

[1252] Data processing: Generating notification messages

[1253] Output: Notification to local government

[1254] Step 9:

[1255] Provide feedback

[1256] Input: Dialogue data, emotion data (historical data)

[1257] Operation: Based on the collected data, the device uses a generated AI model to provide necessary feedback to the user.

[1258] Prompt: You seem a little down lately. Is there anything I can help you with?

[1259] Output: Feedback content (text format)

[1260] (Application Example 2)

[1261] Next, we will explain application example 2. In the following explanation, 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."

[1262] Conventional health and mental state monitoring systems have struggled to grasp users' emotional and health states in real time during their daily activities and to take appropriate action quickly based on that information. Furthermore, in public spaces such as stores, there have been challenges in appropriately understanding customers' emotional states and providing services based on that understanding. The present invention aims to solve these problems and provide a system that can evaluate the health and mental state of users and customers in real time and enable a rapid response when an abnormality is detected.

[1263] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1264] In this invention, the server includes means for authenticating users, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for monitoring the emotional and health states of customers in real time within the store, means for notifying store staff of the results of the customer emotion analysis and health evaluation, and means for providing customer service feedback according to the customer's state. This enables real-time monitoring of the emotional and health states of users and customers, and appropriate responses and feedback based on these.

[1265] Definitions of important words

[1266] "User authentication" is a means by which a system verifies the user's identity.

[1267] Natural language processing is a technology that enables computers to understand, generate, and manage human language.

[1268] "Dialogue data" refers to digital data that includes the content of conversations between users and systems.

[1269] "Health status" refers to information about the user's physical condition and health.

[1270] "Mental state" refers to the user's psychological or emotional state.

[1271] A "database" is a storage system for systematically saving analysis results and data.

[1272] Anomaly detection is the process of detecting unusual states or behaviors.

[1273] "Notification" refers to the act of a system informing users or administrators of specific information.

[1274] "Inside the store" refers to the physical commercial space.

[1275] "Customer" refers to an individual who uses a store or service.

[1276] Real-time monitoring is the process of collecting and analyzing information and data without delay.

[1277] "Emotional analysis" is a technology that evaluates an individual's emotional state through the analysis of voice and video.

[1278] "Health assessment results" refer to the analysis results regarding the health status of users and customers.

[1279] "Feedback" is the act of a system providing responses or advice to a user.

[1280] "Customer service" refers to the process of providing product descriptions, assistance, and customer service to customers.

[1281] Modes for carrying out the invention

[1282] This invention provides a system for routinely monitoring the health and mental state of users and customers and for rapidly detecting abnormalities. A key feature is its integration with an emotion engine, which allows for the evaluation of the emotional state of users and customers. Specific embodiments of this system are described below.

[1283] This system is used in commercial spaces (such as stores) where users and customers are present, allowing store employees to monitor customers' emotional and health states in real time using smart glasses. The main components of the system are as follows:

[1284] 1. User Authentication Method: The system uses facial recognition technology to verify the identity of users and customers. Facial recognition software will utilize libraries such as OpenCV.

[1285] 2. Natural Language Processing (NLP) Methods: The system uses NLP libraries such as Transformers and the BERT model to generate and manage conversations with users and customers. This allows for the analysis of conversation content and the evaluation of health and mental state.

[1286] 3. Emotion Analysis Methods: An emotion engine is used to analyze audio and video data and evaluate the customer's emotional state. This includes facial expression analysis and voice analysis.

[1287] 4. Database: A storage system is used to save the analyzed results. Data on customers' emotional and health states is accumulated, and an anomaly detection algorithm is applied if an anomaly is detected.

[1288] 5. Notification method: If an anomaly is detected, the server will send a notification to the store staff using a communication service such as Twilio.

[1289] 6. Real-time monitoring method: Smart glasses monitor the customer's emotional state and health status in real time and provide feedback to the store staff.

[1290] By combining the above components, it is possible to monitor the health and mental state of users and customers on a daily basis and to quickly notify them if any abnormalities are detected.

[1291] Usage example

[1292] As a concrete example, consider a scenario where a store monitors the health and emotional state of its customers. The store staff wear smart glasses. When a customer enters the store to look at merchandise, facial recognition technology identifies the customer's face, and a dialogue using natural language processing technology begins. The smart glasses analyze the customer's emotional state in real time, and data indicating that "the customer is experiencing stress" is obtained. If appropriate customer service is needed, an alert is sent to the staff via Twilio, allowing them to respond quickly and appropriately.

[1293] Example of a prompt

[1294] "Imagine a customer facial recognition system for emotion analysis. Develop an application that detects customer stress and distress, and supports appropriate feedback and responses."

[1295] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1296] Program processing steps

[1297] Step 1:

[1298] The smart glasses activate the camera and capture images of customers inside the store.

[1299] (Specific action)

[1300] The smart glasses' camera captures video in real time, and that video data is sent to the system.

[1301] (Input) Customer's real-time video data

[1302] (Output) Face feature data required for face recognition

[1303] Step 2:

[1304] The server uses facial recognition technology to identify the customer's face.

[1305] (Specific action)

[1306] The server uses the OpenCV library to perform face recognition on video data and extract customer facial features.

[1307] (Input) Facial feature data

[1308] (Output) Recognized face location information and features

[1309] Step 3:

[1310] The server activates a natural language processing model and generates and manages interactions with the user.

[1311] (Specific action)

[1312] The server uses the Transformers library and the BERT model to analyze conversations between store employees and customers in real time.

[1313] (Input) Text data of conversations with customers

[1314] (Output) Analyzed emotional and health assessment results

[1315] Step 4:

[1316] The server uses an emotion engine to analyze the customer's emotional state.

[1317] (Specific action)

[1318] The server analyzes audio and video data to assess the customer's emotional state in real time. This includes voice tone analysis and facial expression analysis.

[1319] (Input) Voice data, facial expression data

[1320] (Output) Customer's emotional state

[1321] Step 5:

[1322] The server saves the emotion analysis results to a database and detects anomalies.

[1323] (Specific action)

[1324] The server records the emotion analysis results in a database and applies an algorithm to detect consecutive abnormal emotional states.

[1325] (Input) Sentiment analysis result data

[1326] (Output) Anomaly flag and its cause data

[1327] Step 6:

[1328] The server will send a notification based on the anomaly detection results.

[1329] (Specific action)

[1330] The server uses Twilio to send notifications to store employees informing them of unusual customer behavior.

[1331] (Input) Anomaly detection flag

[1332] (Output) Notification message to store staff

[1333] Step 7:

[1334] Smart glasses provide feedback.

[1335] (Specific action)

[1336] The smart glasses display information about the customer's emotional state and hints for how to respond, providing feedback to the store staff.

[1337] (Input) Customer sentiment assessment data, health assessment data

[1338] (Output) Feedback content

[1339] Step 8:

[1340] Store staff will provide appropriate service to customers within the store.

[1341] (Specific action)

[1342] Store staff provide appropriate service and support to customers based on feedback from smart glasses.

[1343] (Input) Feedback content

[1344] (Output) Customer service and support

[1345] The above outlines the specific processing steps for implementing the invention.

[1346] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1347] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1348] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1349] [Fourth Embodiment]

[1350] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1351] As shown in Figure 7, the 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.

[1352] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1353] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1354] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1355] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1356] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1357] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1358] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1359] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1360] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1361] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1362] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1363] This invention provides a system that offers daily support to lonely elderly people and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[1364] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. It also includes means for encrypting interaction data and sending it to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, and means for providing feedback based on the interaction logs.

[1365] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[1366] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" The user's responses are recorded by the device and converted into text. This text data is then encrypted and sent to the server.

[1367] The server analyzes the received data and uses natural language processing to extract keywords and perform sentiment analysis. Based on the analysis results, it evaluates the user's health and mental state. The analysis results are recorded in a database.

[1368] The results stored in the database are constantly monitored by an algorithm designed to detect anomalies. If an anomaly is detected, the server automatically sends a notification to the local government, issuing a warning via email or SMS. This notification includes the username, the date and time the problem occurred, and a brief analysis.

[1369] Furthermore, the device provides feedback to the user based on the collected conversational data. For example, the device generates feedback to reconfirm the user's mental state, such as, "You seem a little down lately. Is there anything bothering you?"

[1370] As a concrete example, consider a scenario where User A interacts with the device daily. User A replies, "Yes, I slept well," and then the device asks, "What are your plans for today?" User A replies, "I don't have any particular plans for today." These responses are sent to the server, and the system evaluates User A's health as good. However, if User A shows lethargic responses a few days later, the server detects an abnormality in their mental state and sends a notification to the local government. In this way, abnormalities can be detected early, enabling appropriate intervention.

[1371] As described above, the present invention provides an effective system for preventing lonely deaths and the progression of dementia by routinely monitoring the health and mental state of lonely elderly people and detecting abnormalities early.

[1372] The following describes the processing flow.

[1373] Step 1:

[1374] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[1375] Step 2:

[1376] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[1377] Step 3:

[1378] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[1379] Step 4:

[1380] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[1381] Step 5:

[1382] The device encrypts the recorded text-based conversation data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[1383] Step 6:

[1384] The server decrypts the received encrypted data and uses natural language processing (NLP) techniques to extract keywords and perform sentiment analysis. Based on the user's responses, the server evaluates their health and mental state.

[1385] Step 7:

[1386] The server records the analysis results in a database. The stored data includes the date, time, user response, and analysis results.

[1387] Step 8:

[1388] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[1389] Step 9:

[1390] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[1391] Step 10:

[1392] The device provides feedback to the user. Based on past conversation data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything I can help you with?"

[1393] This processing step allows for effective monitoring of the health and safety of elderly people living alone, and enables a rapid response if any abnormalities are detected.

[1394] (Example 1)

[1395] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1396] In modern society, it is crucial to detect abnormalities in the health and mental state of lonely elderly individuals early and provide appropriate support. However, conventional systems rely on elderly individuals reporting abnormalities themselves, making early detection difficult. Furthermore, the collection and analysis of conversational data is insufficient, making it difficult to accurately monitor changes in individual health and mental states.

[1397] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1398] In this invention, the server includes means for the terminal to automatically start up at a specified time each morning and authenticate the user by scanning their face; means for generating and managing conversations with the user using natural language processing; means for recording the user's voice responses, converting them to text format and encrypting them; means for transmitting the encrypted text data to the server; means for the server to analyze the received data and perform keyword extraction and sentiment analysis using natural language processing; means for evaluating the user's health and mental state based on the analysis results and recording it in a database; and means for continuously monitoring using an algorithm for detecting abnormalities and notifying the local government if an abnormality is detected. This makes it possible to monitor the health and mental state of lonely elderly people on a daily basis, detect abnormalities early, and take appropriate action.

[1399] A "terminal" is an electronic device that functions as the user interface of a system and manages interactions with the user.

[1400] "User authentication" is a process of verifying a user's identity and is a security function that uses technologies such as facial recognition.

[1401] Natural language processing is a technology that enables computers to understand, interpret, and generate human language, and is used to generate and manage interactions with users.

[1402] "Speech recognition" is a technology that converts speech into text data, and is a technology for processing a user's verbal responses as written information.

[1403] "Encryption" is the process of hiding the contents of data in order to transfer it securely, and it is carried out using a specific algorithm.

[1404] A "server" is a centralized computer system that analyzes received data and stores and manages the results.

[1405] "Data analysis" is the process of processing collected data to derive useful information and insights, and it involves techniques such as natural language processing and other analytical methods.

[1406] "Sentiment analysis" is a technology that identifies and classifies emotions from text and audio data, and is used to evaluate a user's mental state.

[1407] A "database" is a system for efficiently storing, searching, and managing data, and is used to record analytical results.

[1408] Anomaly detection is a technique for identifying data that deviates from normal patterns, and is a process for determining whether there is an abnormality in one's health or mental state.

[1409] "Feedback" refers to advice and confirmations that a system provides based on its interaction with the user and analysis results, and includes messages that include reactions and suggestions to the user.

[1410] "Local government notification" is a function that contacts local government agencies when an anomaly is detected, and this is done via email or SMS.

[1411] Modes for carrying out the invention

[1412] This invention provides a system that offers daily support to lonely elderly individuals and detects abnormalities in their health and mental state at an early stage. Specific embodiments of this system are described below.

[1413] System Configuration

[1414] The system consists mainly of the following elements:

[1415] 1. Terminal

[1416] The terminal is an electronic device installed in the homes of elderly people to manage interactions with the user.

[1417] It automatically starts up at a designated time every morning and uses facial recognition technology to authenticate the user.

[1418] Conversations with users are conducted using a natural language processing engine (e.g., OpenAI GPT-3).

[1419] Use speech recognition software (e.g., Google Cloud Speech-to-Text) to convert the user's voice response into text.

[1420] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption) and sent to the server using a secure protocol (e.g., HTTPS).

[1421] 2. Server

[1422] A server is a centralized computer system that analyzes received data and stores and manages the results.

[1423] We use natural language processing engines (e.g., NLTK and spaCy) to extract keywords and perform sentiment analysis to evaluate the user's health and mental state.

[1424] The analysis results are recorded in an SQL database (e.g., MySQL or PostgreSQL) and continuously monitored by anomaly detection algorithms (e.g., machine learning models).

[1425] If an anomaly is detected, the system will notify the local government via email or SMS API (e.g., Twilio).

[1426] Specific operation of the system

[1427] 1. User Authentication

[1428] The device automatically starts up at 8 AM every morning and scans the user's face with its built-in camera. It also authenticates the user using a facial recognition algorithm (e.g., OpenCV or a common facial recognition library).

[1429] 2. Everyday Conversation Session

[1430] Upon successful authentication, the device activates a generative AI and begins a casual conversation with the user.

[1431] For example, questions such as "Did you sleep well last night?" and "What are your plans for today?" are generated.

[1432] 3. Data collection, encryption, and transmission

[1433] The device records the user's voice response and converts it into text data using speech recognition software.

[1434] The converted text data is encrypted and sent to the server using a secure protocol.

[1435] 4. Data Analysis and Anomaly Detection

[1436] The server analyzes the received data and uses natural language processing to evaluate the user's health and mental state.

[1437] The analysis results are recorded in a database and continuously monitored using an anomaly detection algorithm.

[1438] If an anomaly is detected, a notification will be sent to the local government via email or SMS.

[1439] 5. Provide feedback

[1440] The device provides appropriate feedback to the user based on the analysis results. It uses generative AI to generate the feedback content and converts it into speech using speech synthesis software (e.g., Amazon Polly or Google Text-to-Speech).

[1441] For example, feedback such as, "You seem a little down lately. Is there anything I can help you with?" might be provided.

[1442] Specific example

[1443] Example of interaction with User A:

[1444] Every morning at 8:00 AM, the terminal automatically starts up and scans user A's face to perform authentication.

[1445] Upon successful authentication, the device asks, "Did you sleep well last night?", and User A replies, "Yes, I slept very well."

[1446] This response will be converted to text, encrypted, and sent to the server.

[1447] The server analyzes the data and determines that User A's health status is good.

[1448] A few days later, if user A responds with "I'm tired and don't want to do anything," the server will use that data to detect an anomaly and send a notification to the local government.

[1449] The device provides feedback saying, "You seem a little down lately. Is there anything I can help you with?"

[1450] Thus, the present invention provides a system for routinely monitoring the health and mental state of lonely elderly people, detecting abnormalities early, and providing appropriate intervention.

[1451] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1452] Step 1:

[1453] System startup and user authentication

[1454] Input: System startup time, user's face image

[1455] Output: Authentication success / failure result

[1456] The device is set to automatically start up at a specified time every morning (e.g., 8:00 AM).

[1457] After startup, the built-in camera is used to capture an image of the user's face.

[1458] Using facial recognition technology (e.g., OpenCV or a general facial recognition library), this facial image is compared against pre-registered facial data.

[1459] If authentication is successful, a "Authentication successful" message is sent to the system, and the process proceeds to the next step. If it fails, the system will either retry or issue an alert.

[1460] Step 2:

[1461] Starting a daily conversation session

[1462] Input: Authentication success message

[1463] Output: Generated question prompt, user voice response

[1464] After successful authentication, the terminal launches a generated AI model (e.g., OpenAI GPT-3).

[1465] To begin speaking, generate a prompt sentence (e.g., "Did you sleep well last night?").

[1466] This question is output to the user as an audio message, and the user's voice response is collected.

[1467] Step 3:

[1468] Speech-to-text conversion and encryption

[1469] Input: User voice response

[1470] Output: Encrypted text data

[1471] The device records the collected voice responses using its microphone.

[1472] Use speech recognition software (e.g., Google Cloud Speech-to-Text) to convert recorded audio into text data.

[1473] The converted text data is encrypted using an encryption algorithm (e.g., AES encryption).

[1474] Create encrypted text data and prepare to proceed to the next step.

[1475] Step 4:

[1476] Sending data to the server

[1477] Input: Encrypted text data

[1478] Output: Encrypted data sent to the server

[1479] The device sends encrypted text data to the server using a secure protocol (e.g., HTTPS).

[1480] The server checks the received data and prepares for the next analysis step.

[1481] Step 5:

[1482] Data Analysis

[1483] Input: Encrypted data sent to the server

[1484] Output: Analysis results (assessment of health and mental state)

[1485] The server decrypts the received encrypted data to obtain the text data.

[1486] We will analyze text data using a natural language processing engine (e.g., NLTK or spaCy). Specifically, we will perform keyword extraction and sentiment analysis.

[1487] Based on the analysis results, data is generated to evaluate the user's health and mental state. The evaluation results are then used in the next step.

[1488] Step 6:

[1489] Database recording of analysis results and anomaly detection.

[1490] Input: Analysis results

[1491] Output: Analysis results recorded in the database, anomaly detection notifications.

[1492] The server records the analysis results in an SQL database (e.g., MySQL or PostgreSQL).

[1493] Anomaly detection algorithms (e.g., machine learning models) are used to continuously monitor the analysis results stored in the database.

[1494] If an anomaly is detected, the notification system is automatically activated and sends notifications to relevant parties via email or SMS API (e.g., Twilio).

[1495] Step 7:

[1496] Provide feedback

[1497] Input: Analysis results and evaluation of anomaly detection

[1498] Output: Feedback provided to the user

[1499] The device uses a generative AI model to generate appropriate feedback (e.g., "You seem a little down lately. Is there anything I can help you with?").

[1500] Feedback generated using text-to-speech software (e.g., Amazon Polly or Google Text-to-Speech) is converted into speech and provided to the user.

[1501] The above describes the specific processing flow of this system's program.

[1502] (Application Example 1)

[1503] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1504] In modern society, monitoring the loneliness and health status of the elderly is a critical issue. In particular, the number of elderly people who are unable to receive prompt and appropriate support is increasing. Such situations can lead to serious problems such as deteriorating health and lonely deaths, necessitating effective countermeasures. Furthermore, there is a need for systems used daily by the elderly that efficiently monitor their health and mental state and respond quickly when abnormalities are detected.

[1505] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1506] In this invention, the server includes means for authenticating the user, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for performing user authentication and health checks using facial recognition technology and voice interaction, means for performing sentiment analysis of user responses using natural language processing, and means for automatically detecting anomalies and providing notifications based on the sentiment analysis results. This makes it possible to routinely monitor the health and mental state of elderly people and to quickly notify them if an anomaly is detected.

[1507] "Means of user authentication" refers to a system that uses cameras or biometric authentication technology to identify users and verify their identity.

[1508] "Means for generating and managing user interactions using natural language processing" refers to technologies that enable smooth conversations by allowing computers to understand natural language and generate and manage responses.

[1509] "Methods for analyzing dialogue data to evaluate a user's health and mental state" refers to algorithms that analyze dialogue content to diagnose and evaluate a user's health and mental state.

[1510] "A means of saving analysis results to a database and notifying when an anomaly is detected" refers to a system that stores the results of data analysis in a database and notifies when an anomaly is discovered.

[1511] "A means of performing user authentication and health checks using facial recognition technology and voice dialogue" refers to a system that utilizes facial recognition and voice dialogue technology to authenticate users and verify their health status.

[1512] "Means for performing sentiment analysis of user responses using natural language processing" refers to a technology that uses natural language processing techniques to analyze user responses and evaluate their emotional state.

[1513] "A means of automatically detecting anomalies and providing notifications based on sentiment analysis results" refers to a system that uses the results of sentiment analysis to detect anomalies and automatically provides notifications as needed.

[1514] "Regularly monitoring the user's health and mental state and promptly notifying them if an abnormality is detected" means constantly monitoring the user's health and mental state and immediately notifying relevant parties when an abnormality is discovered.

[1515] This invention is a system that routinely monitors the health and mental state of elderly individuals and promptly notifies them if any abnormalities are detected. Specific embodiments of this system are described below.

[1516] 1. System Configuration

[1517] The system consists of the following main elements:

[1518] User authentication method (facial recognition technology)

[1519] A means of generating and managing user interactions using natural language processing.

[1520] A method for analyzing dialogue data to evaluate the user's health and mental state.

[1521] A means of saving analysis results to a database and notifying when an anomaly is detected.

[1522] A method for user authentication and health checks using facial recognition technology and voice interaction.

[1523] A method for performing sentiment analysis of user responses using natural language processing.

[1524] A method for automatically detecting anomalies and sending notifications based on sentiment analysis results.

[1525] 2. Use of Hardware and Software

[1526] hardware

[1527] Camera (OpenCV compatible)

[1528] Microphone (compatible with speech_recognition library)

[1529] software

[1530] OpenCV: Uses facial recognition technology to authenticate users.

[1531] speech_recognition: Enables voice interaction and checks the user's health status.

[1532] nltk: Uses natural language processing to analyze user response text and perform sentiment analysis.

[1533] requests: Send analysis results to the remote server.

[1534] 3. Data Processing and Analysis

[1535] Dialogue data generated within the system is converted into text using speech recognition technology and further analyzed using natural language processing. During this process, sentiment analysis is performed to assess the user's health and mental state. This evaluation data is stored in a database, and if an anomaly is detected, the system automatically notifies the local government and family.

[1536] 4. Specific Examples

[1537] Example 1: User authentication and health check

[1538] The user is authenticated by facial recognition technology when they stand in front of the camera. After authentication, the system asks aloud, "Did you sleep well last night?", to which the user replies, "Yes, I slept well." This response is converted into text by speech recognition technology and sent to a server for analysis.

[1539] Example 2: Sentiment analysis and notification

[1540] If a user responds with "I don't feel like doing anything in particular today," a negative score is obtained through sentiment analysis using natural language processing. Based on this analysis, the server detects an anomaly and sends notifications to local authorities and family members.

[1541] Examples of prompts for generative AI models

[1542] Use the following text to analyze the user's health status and calculate their emotional score:

[1543] "I slept well last night."

[1544] In this way, the system can routinely monitor the health and mental state of elderly individuals and respond quickly in the event of an emergency. This enables effective interventions to prevent lonely deaths and the progression of dementia.

[1545] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1546] Step 1:

[1547] The device automatically starts up at a designated time each morning and scans the user's face using its camera. The input is the camera image, and the output is the result of user authentication. User authentication is performed using facial recognition technology (OpenCV).

[1548] Step 2:

[1549] If user authentication is successful, the device will prompt for voice input and begin checking the user's health status. It will ask questions such as, "Did you sleep well last night?" The input for this step is the user's voice response, and the output is audio data.

[1550] Step 3:

[1551] The device uses speech recognition technology (speech_recognition library) to convert the user's voice responses into text data. The input is voice data, and the output is text data.

[1552] Step 4:

[1553] Text data is sent from the terminal to the server in an encrypted state. The input here is text data, and the output is encrypted data. The requests library is used for sending.

[1554] Step 5:

[1555] The server analyzes the received text data using natural language processing (NLP) techniques (the nltk library). The input is encrypted text data, and the output is the sentiment analysis result. Specifically, it calculates a sentiment score from the text data.

[1556] Step 6:

[1557] The server evaluates the user's health and mental state based on the analysis results and saves the results to a database. The input is the emotion analysis results, and the output is the operation to save them to the database.

[1558] Step 7:

[1559] The server periodically monitors the database and sends notifications to local governments and families if an anomaly is detected. The input for this step is the analysis results from the database, and the output is a notification email or SMS.

[1560] Step 8:

[1561] The device provides feedback to the user. It generates feedback in the form of, "You seem a little down lately. Is there anything I can help you with?" The input for this step is the result of sentiment analysis, and the output is voice feedback to the user.

[1562] Step 9:

[1563] The prompt text used as input to the generative AI model is: "Analyze the user's health status using the following text and calculate a sentiment score: 'I slept well last night.'" The input is the prompt text, and the output is the generated sentiment score.

[1564] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1565] This invention relates to a system for routinely monitoring a user's health and mental state and rapidly detecting abnormalities. It is particularly characterized by its integration with an emotion engine, which allows the system to also evaluate the user's emotional state. Specific embodiments of this system are described below.

[1566] The system includes means for user authentication, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected. In addition, it includes means for encrypting and sending interaction data to a server, means for authenticating the user using facial recognition, means for notifying local governments via email or SMS if an anomaly is detected, means for providing feedback based on the interaction logs, and an emotion engine that recognizes the user's emotions.

[1567] The emotion engine analyzes audio and image data collected during interactions with the user to identify the user's emotional state. Specifically, the emotion engine uses algorithms that estimate emotions from the user's voice tone, speaking style, facial expressions, and other factors.

[1568] The device automatically starts up at a designated time each morning and performs the user authentication process. User authentication uses facial recognition technology, scanning the user's face with a camera. Upon successful authentication, the device activates a generative AI and begins a casual conversation session with the user.

[1569] The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?" In addition, an emotion engine analyzes the user's emotions in real time from their responses, analyzing the tone of voice and facial expressions when the user answers, "Yes, I slept well." The analysis results are converted into text format as emotion data, encrypted along with the dialogue data, and then sent to the server.

[1570] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. Specifically, both the user's responses and emotional state are considered in the health assessment. For example, even if a user answers "I slept well," if their tone of voice or facial expression is listless, they will be evaluated as mentally exhausted.

[1571] The analysis results are recorded in a database. The data includes the date, time, user response, emotional state, and analysis results. The database is constantly monitored, and an algorithm for detecting anomalies flags consecutive lethargic responses or emotional disturbances as abnormal. If an anomaly is detected, the server automatically notifies the local government via email or SMS to promptly report the user's status.

[1572] Furthermore, the device provides feedback based on the collected conversation and emotional data. For example, if a user is consistently lethargic, the device will check on the user's current situation by saying something like, "You seem a little down lately. Is there anything I can help you with?" and then provide necessary advice and support.

[1573] As a concrete example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if a mental abnormality is detected by the anomaly detection algorithm, a notification is sent to the local government. Subsequently, the local government promptly visits User B's home and takes appropriate action.

[1574] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, achieving more accurate monitoring of health and mental state. As a result, it provides an effective system that improves the quality of life for lonely elderly people and prevents lonely deaths and the progression of dementia.

[1575] The following describes the processing flow.

[1576] Step 1:

[1577] The device automatically starts up at a specified time every morning (e.g., 7:00). The device greets the user with "Good morning. I look forward to working with you today."

[1578] Step 2:

[1579] The device scans the user's face with its camera and authenticates the user using facial recognition software. If authentication is successful, the user is notified with the message, "Authentication complete." If authentication fails, a voice message prompting the user to try again is displayed, saying, "Authentication failed. Please try again."

[1580] Step 3:

[1581] Once authentication is successful, the device activates a generative AI and begins a casual conversation session with the user. The device uses natural language processing to ask the user questions such as, "Did you sleep well last night?"

[1582] Step 4:

[1583] The user answers the device's questions. For example, they might respond, "Yes, I slept well." The device records this response and converts it to text.

[1584] Step 5:

[1585] During a conversation, the device analyzes the user's voice and facial expressions using an emotion engine. It identifies the user's emotional state from factors such as voice tone, speaking style, and facial expressions.

[1586] Step 6:

[1587] The device encrypts the recorded dialogue and emotion data and sends it to the server. Standard encryption algorithms such as AES (Advanced Encryption Standard) are used for data encryption.

[1588] Step 7:

[1589] The server decrypts the received encrypted data and evaluates the user's health and mental state based on natural language processing (NLP) and sentiment analysis using an emotion engine. The server considers both the user's responses and their emotional state when making the evaluation.

[1590] Step 8:

[1591] The server records the analysis results in a database. The stored data includes the date, time, user response, emotional state, and analysis results.

[1592] Step 9:

[1593] The server constantly monitors the analysis results stored in the database and executes algorithms to detect anomalies. For example, if a user consistently fails to respond or if there is no response for several days, an anomaly is flagged.

[1594] Step 10:

[1595] If an anomaly is detected, the server will automatically notify the local government via email or SMS. The notification will include the user's name, the date and time the problem occurred, and a brief analysis result.

[1596] Step 11:

[1597] The device provides feedback to the user. Based on past conversation data and emotional data, it offers emotional support in the form of messages such as, "You seem a little down lately. Is there anything bothering you?"

[1598] This series of processing steps allows the system, combined with the emotion engine, to monitor the user's health and mental state with greater accuracy and respond quickly when an anomaly is detected.

[1599] (Example 2)

[1600] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1601] In modern society, health problems caused by loneliness and mental stress are becoming increasingly serious. Loneliness and mental fatigue, especially among the elderly, are significant social issues, requiring rapid and accurate monitoring and appropriate responses. However, conventional systems have limited means of evaluating users' health and mental states, and lack emotional analysis, making it difficult to accurately detect users' conditions. Furthermore, insufficient means of detecting abnormalities often led to delays in prompt responses.

[1602] The identification processing performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for authenticating the user, means for generating and managing conversations with the user using natural language processing, means for evaluating the user's health and mental state by analyzing conversation data, voice data, and image data, means for storing the analysis results in a database and notifying when an abnormality is detected, and means for identifying the user's emotional state using an emotion engine. This makes it possible to monitor the user's health and mental state in detail, including their emotional state, and to take a quick and appropriate response.

[1603] "User authentication" refers to the method of verifying a user's identity when they access a system, using technologies such as cameras or facial recognition.

[1604] "Natural language processing" is a technology that enables computers to understand, analyze, and generate human language, and is used as a means to generate and manage interactions with users.

[1605] "Dialogue data" refers to data that records the content of conversations that take place between a user and a system.

[1606] "Voice data" refers to data used to record and analyze the voice spoken by a user.

[1607] "Image data" refers to data used to record and analyze a user's face and facial expressions.

[1608] "Health status" refers to the user's physical and mental health condition.

[1609] "Mental state" refers to an assessment of the user's mental state and emotional stability.

[1610] "Analysis results" refer to evaluation results regarding the user's health and mental state obtained from dialogue data and emotional data.

[1611] A "database" is a data structure used to store analysis results and interaction data so that they can be referenced and analyzed later.

[1612] "Means for notifying when an anomaly is detected" refers to a function that automatically sends alerts or notifications to relevant organizations and individuals when an anomaly is detected from the analysis results.

[1613] An "emotion engine" is a general term for algorithms and technologies used to identify a user's emotional state from their voice and facial expressions.

[1614] This invention provides a system for routinely monitoring a user's health and mental state and for rapidly detecting abnormalities. This system is implemented using the hardware and software described below.

[1615] Hardware and software:

[1616] 1. Terminal:

[1617] Camera (using facial recognition technology)

[1618] Microphone (voice data collection)

[1619] Display (for interacting with the user)

[1620] 2. Server:

[1621] Database (for storing analysis results and dialogue data)

[1622] Natural language processing models (generative AI models, e.g., GPT-3)

[1623] Emotion engine (face recognition and voice analysis)

[1624] Program processing:

[1625] User authentication:

[1626] The device automatically activates at a designated time each morning and uses its camera to scan the user's face. This authentication process utilizes facial recognition technology (e.g., OpenCV or FaceNet).

[1627] Everyday conversation session:

[1628] Upon successful authentication, the device activates a generative AI model and initiates a casual conversation session with the user using natural language processing technology. Examples of specific questions include, "Did you sleep well last night?"

[1629] Emotion analysis:

[1630] During user responses, the emotion engine analyzes voice tone and facial expressions in real time. For example, technologies such as Deep Learning for Audio are used to analyze emotions from audio data, and OpenCV and deep learning models are used to analyze emotions from facial expressions.

[1631] Data encryption and transmission:

[1632] Emotional and conversational data are converted into text format and encrypted using encryption technologies such as AES. The encrypted data is then sent to the server.

[1633] Server-based analysis:

[1634] The server receives encrypted data and performs decryption. It uses natural language processing and an emotion engine to evaluate the user's health and mental state. Specifically, both the response content and emotional state are considered when evaluating health.

[1635] Recording to the database:

[1636] The analysis results are stored in a database, recording the date, time, response content, emotional state, and analysis results. The database is constantly monitored, and algorithms for detecting anomalies monitor for consecutive lethargic responses and emotional disturbances.

[1637] Anomaly detection and notification:

[1638] If an anomaly is detected, the server will automatically notify the local government via email or SMS. Specifically, notifications will be sent using the Twilio API, among other methods.

[1639] Provide feedback:

[1640] Based on the collected data, the device's AI model generates and provides necessary feedback to the user. This feedback includes conversational phrases such as, "You seem a little down lately. Is there anything I can help you with?"

[1641] Specific example:

[1642] For example, consider a scenario where User B interacts with the device on a daily basis. If User B responds with "I didn't sleep well last night," and their voice sounds tired and their facial expression is gloomy, the emotion engine analyzes this data and assesses mental fatigue. The analysis results are recorded in a database, and if the anomaly detection algorithm detects a mental abnormality, a notification is sent to the local government. The local government then promptly visits User B's home and takes appropriate action.

[1643] Examples of prompts for a generative AI model:

[1644] Enter a question such as, "Did you sleep well last night?"

[1645] Thus, by combining an emotion engine, the present invention also evaluates the user's emotional state, enabling more accurate monitoring of health and mental state.

[1646] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1647] Step 1:

[1648] Device startup and user authentication

[1649] Input: Time (specified date and time)

[1650] Operation: The device will automatically start up and turn on the camera at a specified time every morning.

[1651] Data processing: The user's face is scanned using facial recognition technology (e.g., OpenCV or FaceNet) and compared with registered user data.

[1652] Output: User authentication result (success / failure)

[1653] Step 2:

[1654] Starting a daily conversation session

[1655] Input: User authentication result (success)

[1656] Operation: Upon successful authentication, the device launches a generated AI model (e.g., GPT-3) and begins a casual conversation with the user using natural language processing technology.

[1657] Prompt: Good morning. Did you sleep well last night?

[1658] Output: User response (text format)

[1659] Step 3:

[1660] User response collection and sentiment analysis

[1661] Input: User response (voice, facial expression)

[1662] Operation: The emotion engine analyzes the user's voice tone and facial expressions in real time. Deep Learning for Audio is used for audio data, and OpenCV and deep learning models are used for facial expression data.

[1663] Data processing: Convert emotional data (tone, facial expression) to text format.

[1664] Output: Sentiment data (text format)

[1665] Step 4:

[1666] Data encryption and transmission

[1667] Input: Dialogue data, emotion data (text format)

[1668] Operation: Encrypts emotional data and dialogue data using encryption technologies such as AES.

[1669] Data processing: Generation of encrypted data

[1670] Output: Encrypted data

[1671] Step 5:

[1672] Decryption of data by the server

[1673] Input: Encrypted data

[1674] Operation: The server receives the transmitted encrypted data and decrypts it using a decryption algorithm such as AES.

[1675] Data processing: Generation of decoded data

[1676] Output: Decoded data (dialogue data, sentiment data)

[1677] Step 6:

[1678] Assessment of physical and mental health

[1679] Input: Dialogue data, emotion data (decoded data)

[1680] Operation: The server uses natural language processing and an emotion engine to evaluate the user's health and mental state. Specifically, it analyzes the response using NLP technology and combines it with emotion data to perform a comprehensive evaluation.

[1681] Data processing: Generation of health status assessments

[1682] Output: Assessment results of health and mental state

[1683] Step 7:

[1684] Database recording and anomaly detection

[1685] Input: Assessment results of health and mental state

[1686] Operation: The analysis results are saved to a database. The database is constantly monitored, and an algorithm detects anomalies, monitoring for continuous lethargic responses and emotional disturbances.

[1687] Data processing: Execution of anomaly detection algorithms

[1688] Output: Anomaly detection result (normal / abnormal)

[1689] Step 8:

[1690] Notification to local government

[1691] Input: Anomaly detection result (anomaly)

[1692] Operation: If an anomaly is detected, the server will use the Twilio API or similar tools to notify the local government via email or SMS.

[1693] Data processing: Generating notification messages

[1694] Output: Notification to local government

[1695] Step 9:

[1696] Provide feedback

[1697] Input: Dialogue data, emotion data (historical data)

[1698] Operation: Based on the collected data, the device uses a generated AI model to provide necessary feedback to the user.

[1699] Prompt: You seem a little down lately. Is there anything I can help you with?

[1700] Output: Feedback content (text format)

[1701] (Application Example 2)

[1702] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1703] Conventional health and mental state monitoring systems have struggled to grasp users' emotional and health states in real time during their daily activities and to take appropriate action quickly based on that information. Furthermore, in public spaces such as stores, there have been challenges in appropriately understanding customers' emotional states and providing services based on that understanding. The present invention aims to solve these problems and provide a system that can evaluate the health and mental state of users and customers in real time and enable a rapid response when an abnormality is detected.

[1704] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1705] In this invention, the server includes means for authenticating users, means for generating and managing user interactions using natural language processing, means for analyzing interaction data to evaluate the user's health and mental state, means for storing the analysis results in a database and notifying if an anomaly is detected, means for monitoring the emotional and health states of customers in real time within the store, means for notifying store staff of the results of the customer emotion analysis and health evaluation, and means for providing customer service feedback according to the customer's state. This enables real-time monitoring of the emotional and health states of users and customers, and appropriate responses and feedback based on these.

[1706] Definitions of important words

[1707] "User authentication" is a means by which a system verifies the user's identity.

[1708] Natural language processing is a technology that enables computers to understand, generate, and manage human language.

[1709] "Dialogue data" refers to digital data that includes the content of conversations between users and systems.

[1710] "Health status" refers to information about the user's physical condition and health.

[1711] "Mental state" refers to the user's psychological or emotional state.

[1712] A "database" is a storage system for systematically saving analysis results and data.

[1713] Anomaly detection is the process of detecting unusual states or behaviors.

[1714] "Notification" refers to the act of a system informing users or administrators of specific information.

[1715] "Inside the store" refers to the physical commercial space.

[1716] "Customer" refers to an individual who uses a store or service.

[1717] Real-time monitoring is the process of collecting and analyzing information and data without delay.

[1718] "Emotional analysis" is a technology that evaluates an individual's emotional state through the analysis of voice and video.

[1719] "Health assessment results" refer to the analysis results regarding the health status of users and customers.

[1720] "Feedback" is the act of a system providing responses or advice to a user.

[1721] "Customer service" refers to the process of providing product descriptions, assistance, and customer service to customers.

[1722] Modes for carrying out the invention

[1723] This invention provides a system for routinely monitoring the health and mental state of users and customers and for rapidly detecting abnormalities. A key feature is its integration with an emotion engine, which allows for the evaluation of the emotional state of users and customers. Specific embodiments of this system are described below.

[1724] This system is used in commercial spaces (such as stores) where users and customers are present, allowing store employees to monitor customers' emotional and health states in real time using smart glasses. The main components of the system are as follows:

[1725] 1. User Authentication Method: The system uses facial recognition technology to verify the identity of users and customers. Facial recognition software will utilize libraries such as OpenCV.

[1726] 2. Natural Language Processing (NLP) Methods: The system uses NLP libraries such as Transformers and the BERT model to generate and manage conversations with users and customers. This allows for the analysis of conversation content and the evaluation of health and mental state.

[1727] 3. Emotion Analysis Methods: An emotion engine is used to analyze audio and video data and evaluate the customer's emotional state. This includes facial expression analysis and voice analysis.

[1728] 4. Database: A storage system is used to save the analyzed results. Data on customers' emotional and health states is accumulated, and an anomaly detection algorithm is applied if an anomaly is detected.

[1729] 5. Notification method: If an anomaly is detected, the server will send a notification to the store staff using a communication service such as Twilio.

[1730] 6. Real-time monitoring method: Smart glasses monitor the customer's emotional state and health status in real time and provide feedback to the store staff.

[1731] By combining the above components, it is possible to monitor the health and mental state of users and customers on a daily basis and to quickly notify them if any abnormalities are detected.

[1732] Usage example

[1733] As a concrete example, consider a scenario where a store monitors the health and emotional state of its customers. The store staff wear smart glasses. When a customer enters the store to look at merchandise, facial recognition technology identifies the customer's face, and a dialogue using natural language processing technology begins. The smart glasses analyze the customer's emotional state in real time, and data indicating that "the customer is experiencing stress" is obtained. If appropriate customer service is needed, an alert is sent to the staff via Twilio, allowing them to respond quickly and appropriately.

[1734] Example of a prompt

[1735] "Imagine a customer facial recognition system for emotion analysis. Develop an application that detects customer stress and distress, and supports appropriate feedback and responses."

[1736] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1737] Program processing steps

[1738] Step 1:

[1739] The smart glasses activate the camera and capture images of customers inside the store.

[1740] (Specific action)

[1741] The smart glasses' camera captures video in real time, and that video data is sent to the system.

[1742] (Input) Customer's real-time video data

[1743] (Output) Face feature data required for face recognition

[1744] Step 2:

[1745] The server uses facial recognition technology to identify the customer's face.

[1746] (Specific action)

[1747] The server uses the OpenCV library to perform face recognition on video data and extract customer facial features.

[1748] (Input) Facial feature data

[1749] (Output) Recognized face location information and features

[1750] Step 3:

[1751] The server activates a natural language processing model and generates and manages interactions with the user.

[1752] (Specific action)

[1753] The server uses the Transformers library and the BERT model to analyze conversations between store employees and customers in real time.

[1754] (Input) Text data of conversations with customers

[1755] (Output) Analyzed emotional and health assessment results

[1756] Step 4:

[1757] The server uses an emotion engine to analyze the customer's emotional state.

[1758] (Specific action)

[1759] The server analyzes audio and video data to assess the customer's emotional state in real time. This includes voice tone analysis and facial expression analysis.

[1760] (Input) Voice data, facial expression data

[1761] (Output) Customer's emotional state

[1762] Step 5:

[1763] The server saves the emotion analysis results to a database and detects anomalies.

[1764] (Specific action)

[1765] The server records the emotion analysis results in a database and applies an algorithm to detect consecutive abnormal emotional states.

[1766] (Input) Sentiment analysis result data

[1767] (Output) Anomaly flag and its cause data

[1768] Step 6:

[1769] The server will send a notification based on the anomaly detection results.

[1770] (Specific action)

[1771] The server uses Twilio to send notifications to store employees informing them of unusual customer behavior.

[1772] (Input) Anomaly detection flag

[1773] (Output) Notification message to store staff

[1774] Step 7:

[1775] Smart glasses provide feedback.

[1776] (Specific action)

[1777] The smart glasses display information about the customer's emotional state and hints for how to respond, providing feedback to the store staff.

[1778] (Input) Customer sentiment assessment data, health assessment data

[1779] (Output) Feedback content

[1780] Step 8:

[1781] Store staff will provide appropriate service to customers within the store.

[1782] (Specific action)

[1783] Store staff provide appropriate service and support to customers based on feedback from smart glasses.

[1784] (Input) Feedback content

[1785] (Output) Customer service and support

[1786] The above outlines the specific processing steps for implementing the invention.

[1787] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1788] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1789] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1790] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1791] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. In the upper and lower directions of the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. Also, the upper side of the concentric circles is where "pleasant" emotions are located, and the lower side is where "unpleasant" emotions are located. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1792] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1793] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1794] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1795] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1796] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1797] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1798] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1799] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1800] 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.

[1801] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1802] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1803] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1804] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1805] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1806] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1807] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1808] The following is further disclosed regarding the embodiments described above.

[1809] (Claim 1)

[1810] Means for authenticating users,

[1811] A means for generating and managing user interactions using natural language processing,

[1812] A means of analyzing dialogue data to evaluate the user's health and mental state,

[1813] A means of saving the analysis results to a database and notifying when an anomaly is detected,

[1814] A system that includes this.

[1815] (Claim 2)

[1816] The system according to claim 1, further comprising means for encrypting dialogue data and transmitting it to a server.

[1817] (Claim 3)

[1818] The system according to claim 1, further comprising means for performing authentication using user facial recognition.

[1819] (Claim 4)

[1820] The system according to claim 1, further comprising means for notifying local governments via email or SMS when an anomaly is detected.

[1821] (Claim 5)

[1822] The system according to claim 1, further comprising means for providing feedback based on a log of the conversation.

[1823] "Example 1"

[1824] (Claim 1)

[1825] A device that automatically activates at a designated time every morning and authenticates the user by scanning their face,

[1826] A means for generating and managing user interactions using natural language processing,

[1827] A means of recording the user's voice response, converting it to text format, and encrypting it,

[1828] A means of sending encrypted text data to a server,

[1829] A method for analyzing data received by a server and using natural language processing to extract keywords and perform sentiment analysis,

[1830] A means of evaluating the user's health and mental state based on the analysis results and recording it in a database,

[1831] A means of continuously monitoring using an algorithm to detect anomalies and notifying the local government when an anomaly is detected,

[1832] A system that includes this.

[1833] (Claim 2)

[1834] The system according to claim 1, further comprising means for generating and providing feedback to a user.

[1835] (Claim 3)

[1836] The system according to claim 1, further comprising means for encrypting and transmitting dialogue data.

[1837] (Claim 4)

[1838] The system according to claim 1, further comprising means for performing authentication using facial recognition technology.

[1839] "Application Example 1"

[1840] (Claim 1)

[1841] Means for authenticating users,

[1842] A means for generating and managing user interactions using natural language processing,

[1843] A means of analyzing dialogue data to evaluate the user's health and mental state,

[1844] A means of saving the analysis results to a database and notifying when an anomaly is detected,

[1845] A means of performing user authentication and health checks using facial recognition technology and voice interaction,

[1846] A method for performing sentiment analysis of user responses using natural language processing,

[1847] A means of automatically detecting anomalies and issuing notifications based on the results of sentiment analysis,

[1848] A system that includes this.

[1849] (Claim 2)

[1850] The system according to claim 1, further comprising means for encrypting dialogue data and transmitting it to a server.

[1851] (Claim 3)

[1852] The system according to claim 1, further comprising means for routinely monitoring the health status of elderly persons and notifying local governments and family members of analysis results for early detection of abnormalities.

[1853] "Example 2 of combining an emotion engine"

[1854] (Claim 1)

[1855] Means for authenticating users,

[1856] A means for generating and managing user interactions using natural language processing,

[1857] A means for evaluating the user's health and mental state by analyzing dialogue data, voice data, and image data,

[1858] A means of saving the analysis results to a database and notifying when an anomaly is detected,

[1859] A means for identifying a user's emotional state using an emotion engine,

[1860] A system that includes this.

[1861] (Claim 2)

[1862] The system according to claim 1, further comprising means for encrypting dialogue data and emotion data and transmitting them to a server.

[1863] (Claim 3)

[1864] The system according to claim 1, further comprising means for performing authentication using user facial recognition.

[1865] "Application example 2 when combining with an emotional engine"

[1866] (Claim 1)

[1867] Means for authenticating users,

[1868] A means for generating and managing user interactions using natural language processing,

[1869] A means of analyzing dialogue data to evaluate the user's health and mental state,

[1870] A means of saving the analysis results to a database and notifying when an anomaly is detected,

[1871] A means of monitoring customers' emotional state and health status in real time within the store,

[1872] A means of notifying store employees of the results of customer emotion analysis and health assessment,

[1873] A means of providing feedback on customer service tailored to the customer's situation,

[1874] A system that includes this.

[1875] (Claim 2)

[1876] The system according to claim 1, further comprising means for encrypting dialogue data and transmitting it to a server.

[1877] (Claim 3)

[1878] The system according to claim 1, further comprising means for performing authentication using user facial recognition. [Explanation of symbols]

[1879] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means for authenticating users, A means for generating and managing user interactions using natural language processing, A means of analyzing dialogue data to evaluate the user's health and mental state, A means of saving the analysis results to a database and notifying when an anomaly is detected, A system that includes this.

2. The system according to claim 1, further comprising means for encrypting dialogue data and transmitting it to a server.

3. The system according to claim 1, further comprising means for performing authentication using user facial recognition.

4. The system according to claim 1, further comprising means for notifying local governments via email or SMS when an anomaly is detected.

5. The system according to claim 1, further comprising means for providing feedback based on a log of the dialogue.

Images