system

Abstract

Provide a system. A method for collecting data from a user includes: means for pre-processing the collected data; means for detecting risk factors from the preprocessed data; means for providing an alert to a user based on the detected risk factors; a means for receiving feedback from users; A system that includes a means to learn and improve the system based on feedback.

Description

[Technical Field]

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

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

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

[0004] In modern society, it is extremely important to detect suicide risk factors early and provide support at the appropriate time. However, conventional methods have difficulty assessing an individual's abnormal behavior or mental state in real time, which often delays emergency support. Furthermore, existing systems lack the ability to learn and improve based on feedback, limiting the accuracy of the support they can provide. To address these issues, a system is needed that can collect and analyze user data in real time and detect risk factors early. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing the following means. Specifically, the system includes a means for collecting various data, such as users' text messages, social media posts, and sensor data. The system then includes a means for preprocessing the collected data and filtering out incomplete data and noise. The system further includes a means for inputting the preprocessed data into a natural language processing model to detect risk factors. The system has a means for providing an alert to the user based on the risk factors and includes a means for receiving feedback from the user. Finally, the system includes a means for learning and improving the system based on the feedback. This makes it possible to detect abnormal user behavior or mental states early and provide appropriate and timely support.

[0006] "User" means an individual who uses this system and provides data.

[0007] "Data Collection Instruments" refers to devices and methods that collect user text messages, social media posts, and sensor data.

[0008] "Data pre-processing means" refers to devices or methods that filter incomplete data and noise in order to convert collected data into a form that is easier to analyze.

[0009] A "risk factor detection means" is a device or method that analyzes preprocessed data and detects factors associated with suicide risk.

[0010] The "alert providing means" refers to a device or method that alerts the user based on the detected risk factors.

[0011] The "feedback receiving means" refers to a device or method for receiving opinions and evaluations from users.

[0012] "Learning and improvement means" refers to devices and methods for improving system performance based on user feedback.

[0013] A "natural language processing model" is an algorithm that analyzes text data and understands the meaning and sentiment of language. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

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

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] MODE FOR CARRYING OUT THE INVENTION

[0036] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[0037] How we collect your data

[0038] Users download the app, create a new account, and then the app asks them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this data and sends it to a server.

[0039] Data Preprocessing

[0040] The server stores the received data in temporary storage. Then, it filters out incomplete data and noise, tokenizes the text data, and uses natural language processing (NLP) libraries to perform morphological analysis of the data and extract useful information.

[0041] Risk factor detection

[0042] The server inputs the preprocessed data into a natural language processing model. This model performs keyword and sentiment analysis to detect suicide risk factors. It also analyzes the sensor data to detect abnormal patterns. From this, the server scores the overall risk and generates an alert if the score exceeds a certain threshold.

[0043] Providing alerts

[0044] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[0045] Gathering feedback and training the system

[0046] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[0047] Specific examples

[0048] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[0049] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[0050] The processing flow will be explained below.

[0051] Step 1:

[0052] Users download the app and create a new account, entering basic information like name, age, and gender, and granting access to text messages, social media posts, and sensor data.

[0053] Step 2:

[0054] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[0055] Step 3:

[0056] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[0057] Step 4:

[0058] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[0059] Filter out incomplete data and noise.

[0060] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[0061] Step 5:

[0062] The server inputs the preprocessed data into a natural language processing model, which performs the following analysis:

[0063] Keyword analysis: Detect specific risky keywords.

[0064] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[0065] Step 6:

[0066] The server analyzes the sensor data to detect abnormal behavioral patterns and lifestyle changes, including:

[0067] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[0068] Decreased activity level

[0069] Step 7:

[0070] The server combines the scores generated from the text data and sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[0071] Step 8:

[0072] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[0073] A message expressing concern

[0074] Chat link with a counselor

[0075] Emergency contact information

[0076] Step 9:

[0077] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[0078] Step 10:

[0079] The user inputs feedback into the terminal on whether the alert and the provided assistance were appropriate, including an evaluation of the timing and content of the alert.

[0080] Step 11:

[0081] The device sends the collected feedback to a server, which receives it and uses the feedback data to retrain the AI ​​model, improving the system's performance and providing more accurate alerts and assistance in the future.

[0082] As described above, this system collects and analyzes user data and provides support at the appropriate time, thereby detecting risk factors for suicide early and helping to save lives.

[0083] Example 1

[0084] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0085] Suicide prevention is an important issue in modern society. However, conventional methods currently struggle to detect a user's psychological state early and provide appropriate support in a timely manner. In particular, it is technically difficult to detect psychological risk factors early from personal text messages, social media posts, and daily activity data. Therefore, there is a need for a system that can monitor a user's mental health state in real time and provide necessary support promptly.

[0086] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0087] In this invention, the server includes means for collecting information from the user, means for temporarily storing the information, means for cleansing and tokenizing the information, means for analyzing the information using natural language processing, means for detecting risk factors and calculating a risk score, means for providing a warning, means for receiving feedback, and means for learning and improving the system, thereby enabling early detection of the user's psychological state and providing appropriate support in a timely manner.

[0088] "User" refers to an individual who uses this system.

[0089] "Information" refers to data collected from users, including text messages, social media posts, and activity data.

[0090] "Means of collection" refers to the functionality of an application or device to obtain information from a user.

[0091] "Temporary storage means" refers to a storage system or database for temporarily storing collected information.

[0092] "Cleansing and tokenization methods" refers to technologies that remove incomplete data and noise from information and perform processing to segment and identify text data.

[0093] "Means of analysis using natural language processing" refers to the process of analyzing information using natural language processing techniques such as morphological analysis and sentiment analysis.

[0094] "Means for detecting risk factors and calculating a risk score" refers to the function of identifying elements that indicate suicide risk from the analysis results and calculating a comprehensive assessment of this as a score.

[0095] "Means for providing warnings" refers to a function that notifies the user of warnings and support information when a risk factor is detected.

[0096] The "means for receiving feedback" refers to a method for collecting feedback from the user as to whether the assistance provided was appropriate.

[0097] "Means of learning and improving the system" refers to the ability to retrain AI models based on feedback and improve the overall performance and accuracy of the system.

[0098] This invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. The system collects various information from users and provides alerts based on the results of analyzing that information. It also collects feedback from users and learns and improves the system's accuracy.

[0099] How we collect your information

[0100] Users download the app, create a new account, and the app prompts them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this information and sends it to a server.

[0101] Information preprocessing

[0102] The server stores the received information in temporary storage. Then, the server filters out incomplete data and noise, tokenizes the text information, and uses a natural language processing (NLP) library (e.g., SpaCy or NLTK) to perform morphological analysis of the information and extract useful information.

[0103] Risk factor detection

[0104] The server inputs the preprocessed information into natural language processing models (e.g., BERT or GPT-3 (registered trademark)). These models analyze keywords and sentiment to detect suicide risk factors. They also analyze sensor data to detect abnormal patterns. From this, the server scores the overall risk and generates an alert if the score exceeds a certain threshold.

[0105] Providing alerts

[0106] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[0107] Gathering feedback and training the system

[0108] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[0109] Specific examples

[0110] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[0111] Example prompts to input to a generative AI model:

[0112] User 01 sends the following message: "I've been feeling down lately." Conduct a sentiment analysis of this message and assess its risk factors.

[0113] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0114] Step 1:

[0115] A user downloads the app and creates a new account.

[0116] Input: Account information entered by the user (name, email address, password)

[0117] Specific operation: After downloading the application from the app store, the user installs it, then launches the app, enters the required information on the new account creation screen, and presses the "Register" button.

[0118] Output: The account information is sent to the server and the new account is saved in the database.

[0119] Step 2:

[0120] Users consent to providing information and text messages, social media posts, and sensor data are collected.

[0121] Input: User-entered consent to provide information

[0122] Specific operation: The user consents to providing information on the account settings screen. After consent, the device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server.

[0123] Output: The collected information is sent to the server and stored in temporary storage.

[0124] Step 3:

[0125] The server preprocesses the received information.

[0126] Input: User text messages, social media posts, and sensor data sent from the device

[0127] How it works: The server stores the collected information in temporary storage, filters out incomplete data and noise, and then uses a natural language processing library (e.g., SpaCy, NLTK) to tokenize the text information and perform morphological analysis.

[0128] Output: Preprocessed and clean information

[0129] Step 4:

[0130] The server inputs the preprocessed information into a natural language processing model to detect risk factors.

[0131] Input: Preprocessed information

[0132] How it works: The server inputs the preprocessed information into a generative AI model (e.g., GPT-3) to perform sentiment analysis and keyword extraction. It then analyzes the sensor data to detect anomalous patterns. This allows the server to calculate a risk score for each data point.

[0133] Output: Risk factor detection results and risk score

[0134] Step 5:

[0135] If the risk score exceeds a certain threshold, the server generates an alert and sends it to the terminal.

[0136] Input: Risk score

[0137] Specific operation: The server evaluates whether the risk score exceeds the set threshold. If it does, it generates an alert message and sends it to the terminal.

[0138] Output: Alert information

[0139] Step 6:

[0140] The device notifies the user of the alert.

[0141] Input: Alert information sent from the server

[0142] What it does: The device immediately notifies the user of an alert message, which includes a warning message, a chat link with a counselor, and emergency contact information.

[0143] Output: The alert message that is displayed to the user.

[0144] Step 7:

[0145] The user provides feedback and the feedback information is sent to the server.

[0146] Input: User feedback information (e.g., appropriateness of assistance)

[0147] Specific operation: The user provides feedback within the app on whether the provided assistance was appropriate. The device then sends this feedback information to the server.

[0148] Output: Feedback information sent to the server

[0149] Step 8:

[0150] The server uses feedback to learn and improve the system.

[0151] Input: User feedback information

[0152] Specific operation: The server uses the received feedback information to retrain the AI ​​model and improve the accuracy of the entire system, thereby improving the accuracy of the next risk factor detection and alert.

[0153] Output: Improved AI models and increased system accuracy

[0154] (Application example 1)

[0155] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0156] In modern society, mental health issues can increase the risk of suicide, creating a need for early detection and rapid response. However, existing systems lack sufficient means for integrating emotion analysis and biometric sensor data to assess risk, making it difficult to provide support at the appropriate time. The present invention aims to solve this problem by more accurately monitoring a user's mental health status and providing support at the appropriate time.

[0157] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0158] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for detecting risk factors from the preprocessed data, means for providing an alert to the user based on the detected risk factors, means for receiving feedback from the user, means for learning and improving the system based on the feedback, means for evaluating the user's emotions using an emotion analysis model, and means for analyzing biosensor data to detect abnormal patterns. This makes it possible to comprehensively analyze a variety of user data, detect risks early, and provide support at an appropriate time.

[0159] "Means for collecting data from users" refers to devices or software systems for obtaining users' text messages, social media posts, sensor data, etc.

[0160] "Means for pre-processing collected data" refers to processes or algorithms used to organize, filter, tokenize, etc. the data obtained.

[0161] "Risk factor detection tools" are algorithms or models that analyze pre-processed data and assess suicide risk based on specific patterns or keywords.

[0162] An "alert providing means" is a device or system for sending a warning or caution to a user based on a detected risk factor.

[0163] The "means for receiving feedback from users" refers to an interface or data collection system for collecting ratings and comments provided by users.

[0164] "Means of learning and improving the system" refers to the process of updating AI models and algorithms based on collected feedback to improve their accuracy.

[0165] The "means for evaluating emotions using an emotion analysis model" is a system that extracts and evaluates emotions from text data using a natural language processing model or other analysis tools.

[0166] "Means for analyzing biometric sensor data to detect abnormal patterns" refers to algorithms and analytical tools that analyze biometric data such as heart rate and sleep patterns to detect patterns that are out of the ordinary.

[0167] The present invention relates to a system for monitoring a user's mental health status, detecting risks early, and providing support at an appropriate time. The system is composed of the following main components:

[0168] 1. Data collection methods:

[0169] It collects users' text messages, social media posts, and biometric sensor data (e.g., heart rate, sleep patterns) using devices such as smartphones, smart glasses, and smart watches.

[0170] 2. Data preprocessing methods:

[0171] The server stores the collected data in temporary storage. It then filters out incomplete data and noise and tokenizes the text data. It uses a natural language processing library (e.g., TextBlob) to perform morphological analysis of the data and extract useful information. It also filters and cleans the sensor data in the same way.

[0172] 3. Risk factor detection methods:

[0173] The preprocessed data is input into a natural language processing model (e.g., an NLP model). This model analyzes emotions from the text and calculates an emotion score. It also analyzes biometric sensor data to detect abnormal patterns (e.g., high heart rate or abnormal sleep patterns). The server comprehensively evaluates this information and identifies risk factors for suicide.

[0174] 4. Alert Delivery Method:

[0175] If a risk factor is detected, the server will send an alert to the user's device, which will include a warning message, a chat link with a counselor, and emergency contact information, allowing the user to receive immediate and appropriate assistance.

[0176] 5. Feedback collection methods:

[0177] After providing the assistance, the user provides feedback on whether the assistance provided was appropriate, and the terminal transmits this feedback information to the server.

[0178] 6. System Learning and Improvement Measures:

[0179] The server retrains the AI ​​model based on feedback collected from users to improve the accuracy of the system.

[0180] Example

[0181] For example, if User A sends a message saying, "I've been feeling down lately," this message is collected from a smartphone app. The server tokenizes this message and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to User A's smartphone saying, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[0182] Prompt Sentence Examples

[0183] When a user sends a message saying, "I've been feeling down lately," the emotion score detected from this message is 0.7, indicating a very negative state. The user's heart rate data also shows abnormalities, indicating that immediate assistance is required.

[0184] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0185] Step 1:

[0186] A user installs the application and creates a new account. The user's smartphone, smart glasses, or smartwatch becomes a device for collecting text messages, social media posts, and biometric sensor data. This collects user input data (e.g., a text message saying "I've been feeling down lately") and sensor data (e.g., heart rate, sleep patterns).

[0187] Input: User text messages, social media posts, biometric sensor data

[0188] Output: Raw data sent to the server

[0189] Step 2:

[0190] The server stores the collected data in temporary storage. Next, it performs preprocessing, filtering out incomplete data and noise and tokenizing the text data. Specifically, it uses a natural language processing library (e.g., TextBlob) to perform morphological analysis and extract useful information. It also cleans the sensor data in a similar way, removing incomplete values.

[0191] Input: Collected raw data

[0192] Output: Filtered and tokenized preprocessed data

[0193] Step 3:

[0194] The server inputs the preprocessed text data into a natural language processing model (NLP model). The NLP model performs sentiment analysis and calculates a sentiment score for the text (e.g., -0.7). At the same time, it also detects abnormal patterns in the sensor data (e.g., a heart rate above 100). This allows for a comprehensive calculation of suicide risk.

[0195] Input: Preprocessed text data and sensor data

[0196] Output: Sentiment score and abnormal pattern detection results

[0197] Step 4:

[0198] The server evaluates risk factors based on the emotion score and abnormal patterns in the sensor data. If the risk factor is determined to be high (e.g., emotion score -0.7 or heart rate above 100), an alert is generated. The generated alert includes a warning message, a link to counseling services, and emergency contact information.

[0199] Input: Sentiment scores and anomaly pattern detection results

[0200] Output: Generated alerts

[0201] Step 5:

[0202] The server then sends the generated alert to the user's device. The user immediately receives the alert on their smartphone or smart glasses and checks the notification. The notification includes a warning such as, "Your recent messages have caused you concern. Counseling services are available," along with emergency contact information.

[0203] Input: Generated alert

[0204] Output: Notification displayed on the user's device

[0205] Step 6:

[0206] The user provides feedback on whether the provided assistance was appropriate. The terminal sends this feedback information to the server, which uses it to improve the system.

[0207] Input: User feedback

[0208] Output: Feedback information sent to the server

[0209] Step 7:

[0210] The server uses the collected feedback information to retrain the AI ​​model and improve the accuracy of the system, which will improve the accuracy and speed of future risk detection.

[0211] Input: Feedback information

[0212] Output: Retrained AI model

[0213] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0214] MODE FOR CARRYING OUT THE INVENTION

[0215] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects a variety of data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the system's accuracy. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it enables more accurate detection of risk factors. Detailed embodiments of the system are described below.

[0216] How we collect your data

[0217] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[0218] Data Preprocessing

[0219] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[0220] Filter out incomplete data and noise.

[0221] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library to extract useful information.

[0222] Emotion recognition by emotion engine

[0223] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[0224] Categorize emotions as negative, positive, or neutral.

[0225] A score is generated for each emotion category.

[0226] Risk factor detection

[0227] The server comprehensively analyzes the emotion scores output from the emotion engine and the sensor data. Specifically, it performs the following steps:

[0228] Keyword analysis: Detect specific risky keywords.

[0229] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[0230] Behavioral pattern detection: Analyze sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[0231] Calculating the overall risk score

[0232] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[0233] Providing alerts

[0234] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[0235] A message expressing concern

[0236] Chat link with a counselor

[0237] Emergency contact information

[0238] Gathering feedback and training the system

[0239] The user checks the notification and enters feedback into the device on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and enables it to provide more accurate alerts and assistance in the future.

[0240] Specific examples

[0241] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[0242] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[0243] The processing flow will be explained below.

[0244] Step 1:

[0245] Users download the app and create a new account by entering basic information (such as name, age, and gender). They then grant permission for access to text messages, social media posts, and sensor data.

[0246] Step 2:

[0247] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[0248] Step 3:

[0249] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[0250] Step 4:

[0251] The server stores the received data in temporary storage. Then, data preprocessing begins. Specifically, it performs the following processes:

[0252] Filter out incomplete data and noise.

[0253] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[0254] Step 5:

[0255] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[0256] Classify the sentiment of each text message or post as positive, negative, or neutral.

[0257] A score is generated for each emotion category.

[0258] Step 6:

[0259] The server uses natural language processing models to analyze the risk factors in the text data. Specifically, it performs the following tasks:

[0260] Keyword analysis: Detect specific risky keywords.

[0261] Sentiment Analysis: Evaluating the sentiment of text data and integrating the scores generated by the sentiment engine.

[0262] Step 7:

[0263] The server integrates the emotion engine score and the results of risk factor analysis of the text data to calculate an overall risk score, and generates an emergency alert if the score exceeds a certain threshold.

[0264] Step 8:

[0265] The server analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Examples include:

[0266] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[0267] Decreased activity level

[0268] Step 9:

[0269] The server integrates the emotion score, the text data risk score, and the sensor data results to make a final risk assessment. If the overall risk score exceeds a threshold, an alert is generated.

[0270] Step 10:

[0271] The device will notify the user of the alert information received from the server. The notification will be displayed as a popup and will include the following:

[0272] A message expressing concern

[0273] Chat link with a counselor

[0274] Emergency contact information

[0275] Step 11:

[0276] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[0277] Step 12:

[0278] The user inputs feedback into the terminal, including an evaluation of the timing and content of the alert, on whether the alert and the provided assistance were appropriate.

[0279] Step 13:

[0280] The device sends feedback to the server, which receives it and uses it to retrain the AI ​​model, improving the system's accuracy and enabling it to provide better alerts and assistance in the future.

[0281] This is the specific processing flow of this system, which combines an emotion engine, making it possible to grasp the user's emotional state more accurately and provide support at the appropriate time.

[0282] Example 2

[0283] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0284] In recent years, mental health issues such as psychological stress and depression have become a major social issue. In particular, it is important to detect suicide risk factors early and provide support at the appropriate time. However, conventional systems can sometimes have difficulty accurately understanding a user's mental state and intervening quickly. For this reason, there is a demand for a system that can analyze a variety of data from users and detect risk factors with high accuracy.

[0285] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0286] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for inputting the preprocessed data into an emotion analysis engine to generate an emotion score, and means for analyzing the emotion score and sensor data to detect risk factors, thereby enabling a highly accurate analysis of the user's mental state and early detection of risk factors.

[0287] "User" refers to any individual or entity that uses the System.

[0288] "Data" refers to information collected from users, such as text messages, social media posts, and sensor data.

[0289] "Means for collection" refers to hardware and software for collecting data with the user's permission.

[0290] "Preprocessing means" refers to the process by which collected data is filtered, tokenized, and prepared for analysis.

[0291] "Sentiment analysis engine" refers to an algorithm or model for generating sentiment scores based on pre-processed text data.

[0292] "Emotion score" refers to a numerical representation of the intensity and type of emotion extracted from text data.

[0293] "Sensor data" refers to data collected from devices that measure a user's activity patterns and biometric information.

[0294] "Risk factors" refer to certain behaviors or psychological states that indicate a user is at increased risk of suicide.

[0295] An "alert" refers to a message that notifies a user or interested party that a risk factor has been detected.

[0296] "Feedback" refers to the user providing the system with their evaluation and opinion of the assistance provided.

[0297] "Learn and improve" refers to the process of retraining the system's algorithms and models based on feedback to improve performance.

[0298] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[0299] How we collect your data

[0300] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[0301] Data Preprocessing

[0302] The server stores the received data in temporary storage, after which it performs preprocessing on the data.

[0303] Specifically, it filters out incomplete data and noise, tokenizes the text data, and performs morphological analysis using natural language processing (NLP) libraries (e.g., NLTK and spaCy) to extract useful information.

[0304] Emotion recognition by emotion engine

[0305] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google® Cloud Natural Language API), which classifies negative, positive, and neutral sentiment and generates a score for each sentiment category.

[0306] Risk factor detection

[0307] The server comprehensively analyzes the emotion score output from the emotion engine and the sensor data. Specifically, it detects specific risk keywords, evaluates the emotion of the text data, and generates a score: positive, negative, or neutral. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[0308] Calculating the overall risk score

[0309] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score. If this overall risk score exceeds a certain threshold, the server generates an emergency alert.

[0310] Providing alerts

[0311] The device will notify the user of the alert received from the server. The notification will be displayed as a popup and will include the following:

[0312] A message expressing concern

[0313] Chat link with a counselor

[0314] Emergency contact information

[0315] Gathering feedback and training the system

[0316] The user checks the notification and enters feedback on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and allows it to provide more accurate alerts and assistance in the future.

[0317] Specific examples

[0318] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the result is determined to have a high negative score, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[0319] This allows the system to detect the user's mental state early and provide support at the appropriate time.

[0320] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0321] Step 1:

[0322] Users download the app and create a new account, entering basic information such as name, age, and gender, and granting access to text messages, social media posts, and sensor data, which is then sent by the device to a server.

[0323] Input: Name, age, gender, text messages, social media posts, sensor data

[0324] Output: Send basic information and permission information to the server

[0325] Step 2:

[0326] The server receives the user's basic information and permission information and stores it in temporary storage. The device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server. The server stores the received data in temporary storage.

[0327] Input: User basic information, text messages, social media posts, sensor data

[0328] Output: Saving data to temporary storage

[0329] Step 3:

[0330] The server begins preprocessing the collected data by filtering out incomplete data and noise, tokenizing the text data, and extracting useful information from the data by performing morphological analysis using a natural language processing (NLP) library (e.g., NLTK or spaCy).

[0331] Input: Raw data stored in temporary storage

[0332] Output: Preprocessed data (filtered, tokenized, morphologically analyzed)

[0333] Step 4:

[0334] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google Cloud Natural Language API), which generates a sentiment score for the text and classifies it into negative, positive, or neutral categories.

[0335] Input: Preprocessed text data

[0336] Output: Sentiment scores (negative, positive, neutral scores)

[0337] Step 5:

[0338] The server comprehensively analyzes the emotion score and sensor data. Specifically, it detects specific risk keywords and evaluates the emotion score. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Based on the results of this analysis, it generates individual scores.

[0339] Input: emotion scores, sensor data

[0340] Output: Risk factor score

[0341] Step 6:

[0342] The server combines the generated individual scores to calculate an overall risk score, and if this overall risk score exceeds a certain threshold, the server generates an emergency alert and sends it to the device.

[0343] Input: Risk factor score

[0344] Output: Overall risk score, emergency alert

[0345] Step 7:

[0346] The device notifies the user of emergency alerts received from the server. The notification appears as a pop-up and includes a message of concern, a chat link to a counselor, and emergency contact information.

[0347] Input: Emergency Alert

[0348] Output: A popup notification to the user

[0349] Step 8:

[0350] The user checks the notification and inputs feedback on whether the provided assistance was appropriate or not into the terminal, which then transmits this feedback information to the server.

[0351] Input: Feedback

[0352] Output: Sending feedback to the server

[0353] Step 9:

[0354] The server uses the received feedback data to retrain the AI ​​model, improving the system's performance and enabling it to provide more accurate alerts and assistance in the future.

[0355] Input: Feedback data

[0356] Output: Retrained AI model

[0357] (Application example 2)

[0358] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0359] In modern society, it is increasingly important to maintain users' mental health, detect risk factors early, and provide appropriate support. However, conventional systems have had difficulty properly collecting and analyzing diverse data and detecting risk factors in real time. Furthermore, a lack of means to provide support at the appropriate time has made it difficult to intervene effectively to ensure users' mental health. Therefore, the present invention aims to solve these problems by detecting users' mental states early and providing appropriate support.

[0360] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0361] In this invention, the server includes means for collecting data from the user, means for preprocessing the collected data, and means for detecting risk factors from the preprocessed data. This allows for the use of emotion scores and behavioral patterns in detecting risk factors. Furthermore, by using a means for providing notifications via a smartphone or smart glasses, it is possible to provide prompt and accurate alerts to the user. Furthermore, the system can receive feedback from the user and continue to learn and improve based on that feedback, thereby enabling continuous monitoring of the user's mental health and providing appropriate support.

[0362] "User" means an individual or organization that uses this system.

[0363] A "means for collecting data" is a device or function for transmitting user text messages, social media posts, sensor data, etc. to a server.

[0364] A "preprocessing means" is a device or function for organizing collected data, filtering out incomplete data and noise, and converting it into an analyzable format.

[0365] The "means for detecting risk factors" is a device or function for evaluating the user's mental health status and behavioral patterns from the preprocessed data and recognizing abnormal conditions and risks.

[0366] A "means for providing an alert" is a device or function for providing a warning or counseling suggestion to a user when a risk factor is detected.

[0367] A "means for receiving feedback" is a device or function for informing the server how the user responded to the alert or whether the assistance provided was adequate.

[0368] A "means for learning and improving the system" is a device or function that updates the system's algorithms based on user feedback to improve accuracy and effectiveness.

[0369] An "emotion score" is a numerical assessment of a user's emotional state based on text data and other data.

[0370] "Behavioral patterns" refer to a user's daily behavior and lifestyle habits analyzed from sensor data.

[0371] A "smartphone" is a portable communication device that has functions such as voice calls, text message sending, and internet connection.

[0372] "Smart glasses" are eyeglass-type devices incorporating display devices and sensors, and are devices that provide visual information to users.

[0373] A system for implementing this invention mainly uses the following hardware and software. The hardware uses a smartphone or smart glasses. The software includes a natural language processing (NLP) library, an emotion recognition engine, and cloud data storage. For example, "spaCy" or "NLTK" can be used as the natural language processing library. Possible emotion recognition engines include "Amazon Comprehend" or "Google Cloud Natural Language API." Furthermore, "AWS (registered trademark) S3" or "Google Cloud Storage" is used as the cloud data storage. "React Native" or "Flutter (registered trademark)" is used for mobile app development.

[0374] The server collects user data such as text messages, social media posts, and sensor data. During data preprocessing, the collected data is organized and incomplete data and noise are filtered. The preprocessed data is input into a natural language processing model and an emotion recognition engine. The server uses emotion scores and behavioral patterns to detect risk factors and provide alerts to users.

[0375] As a concrete example, a user may send a message via a smartphone app saying, "I'm feeling depressed." The server tokenizes the message and analyzes it using a natural language processing model and an emotion recognition engine. If the text is determined to indicate a high risk factor, the server will provide an alert to the user via their smartphone or smart glasses. The alert message may say, "Your recent messages have caused us concern. Counseling services are available."

[0376] The prompt statement in the example program is:

[0377] Input prompt: User types "I'm feeling down"

[0378] System Output: "Your recent messages have me concerned. Counseling services are available."

[0379] This allows users to receive appropriate assistance quickly. The system also learns and improves based on user feedback, which will enable more accurate risk factor detection and assistance provision in the future.

[0380] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0381] Step 1:

[0382] A user uses a smartphone or smart glasses to input a text message, for example, the user inputs the message "I'm feeling down," which is sent to the system.

[0383] Step 2:

[0384] The terminal sends the entered text message to the server, where it is temporarily stored.

[0385] Step 3:

[0386] The server preprocesses the received text message. The input is the text message, and the output is tokenized text. Specifically, it uses a natural language processing library (e.g., spaCy) to tokenize the message and perform morphological analysis.

[0387] Step 4:

[0388] The server inputs the preprocessed text data into an emotion recognition engine. The input is tokenized text, and the output is an emotion score. This process uses services such as Amazon Comprehend or Google Cloud Natural Language API to generate an emotion score (e.g., negative, positive, neutral).

[0389] Step 5:

[0390] The server combines the generated emotion scores with the sensor data to calculate an overall risk score. Specifically, it numerically evaluates and combines the emotion scores and behavioral pattern data (e.g., sleep patterns, activity levels). This results in an overall risk score.

[0391] Step 6:

[0392] If the overall risk score exceeds a certain threshold, the server generates an alert and notifies the device. Specifically, the notification content is created as a message such as "Your recent messages have caused you concern. Counseling services are available." and sent to the smartphone or smart glasses.

[0393] Step 7:

[0394] The device displays the received alert notification to the user, who can then check the alert and access information such as counseling services or emergency contacts as needed.

[0395] Step 8:

[0396] The user inputs feedback about the alert into the terminal, such as "The notified support was appropriate" or "I used the service."

[0397] Step 9:

[0398] The terminal transmits the feedback from the user to the server, which stores the feedback.

[0399] Step 10:

[0400] The server uses the feedback it receives to learn and improve the system. Specifically, it inputs the feedback data into a generative AI model, retraining the algorithm and improving the overall system performance, which will enable more accurate risk factor detection and assistance provision in the future.

[0401] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0402] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0403] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0404] [Second embodiment]

[0405] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0406] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0407] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0408] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0409] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0410] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0411] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0412] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0413] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0414] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0415] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0416] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0417] MODE FOR CARRYING OUT THE INVENTION

[0418] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[0419] How we collect your data

[0420] Users download the app, create a new account, and then the app asks them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this data and sends it to a server.

[0421] Data Preprocessing

[0422] The server stores the received data in temporary storage. Then, it filters out incomplete data and noise, tokenizes the text data, and uses natural language processing (NLP) libraries to perform morphological analysis of the data and extract useful information.

[0423] Risk factor detection

[0424] The server inputs the preprocessed data into a natural language processing model. This model performs keyword and sentiment analysis to detect suicide risk factors. It also analyzes the sensor data to detect abnormal patterns. From this, the server scores the overall risk and generates an alert if the score exceeds a certain threshold.

[0425] Providing alerts

[0426] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[0427] Gathering feedback and training the system

[0428] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[0429] Specific examples

[0430] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[0431] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[0432] The processing flow will be explained below.

[0433] Step 1:

[0434] Users download the app and create a new account, entering basic information like name, age, and gender, and granting access to text messages, social media posts, and sensor data.

[0435] Step 2:

[0436] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[0437] Step 3:

[0438] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[0439] Step 4:

[0440] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[0441] Filter out incomplete data and noise.

[0442] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[0443] Step 5:

[0444] The server inputs the preprocessed data into a natural language processing model, which performs the following analysis:

[0445] Keyword analysis: Detect specific risky keywords.

[0446] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[0447] Step 6:

[0448] The server analyzes the sensor data to detect abnormal behavioral patterns and lifestyle changes, including:

[0449] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[0450] Decreased activity level

[0451] Step 7:

[0452] The server combines the scores generated from the text data and sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[0453] Step 8:

[0454] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[0455] A message expressing concern

[0456] Chat link with a counselor

[0457] Emergency contact information

[0458] Step 9:

[0459] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[0460] Step 10:

[0461] The user inputs feedback into the terminal on whether the alert and the provided assistance were appropriate, including an evaluation of the timing and content of the alert.

[0462] Step 11:

[0463] The device sends the collected feedback to a server, which receives it and uses the feedback data to retrain the AI ​​model, improving the system's performance and providing more accurate alerts and assistance in the future.

[0464] As described above, this system collects and analyzes user data and provides support at the appropriate time, thereby detecting risk factors for suicide early and helping to save lives.

[0465] Example 1

[0466] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0467] Suicide prevention is an important issue in modern society. However, conventional methods currently struggle to detect a user's psychological state early and provide appropriate support in a timely manner. In particular, it is technically difficult to detect psychological risk factors early from personal text messages, social media posts, and daily activity data. Therefore, there is a need for a system that can monitor a user's mental health state in real time and provide necessary support promptly.

[0468] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0469] In this invention, the server includes means for collecting information from the user, means for temporarily storing the information, means for cleansing and tokenizing the information, means for analyzing the information using natural language processing, means for detecting risk factors and calculating a risk score, means for providing a warning, means for receiving feedback, and means for learning and improving the system, thereby enabling early detection of the user's psychological state and providing appropriate support in a timely manner.

[0470] "User" refers to an individual who uses this system.

[0471] "Information" refers to data collected from users, including text messages, social media posts, and activity data.

[0472] "Means of collection" refers to the functionality of an application or device to obtain information from a user.

[0473] "Temporary storage means" refers to a storage system or database for temporarily storing collected information.

[0474] "Cleansing and tokenization methods" refers to technologies that remove incomplete data and noise from information and perform processing to segment and identify text data.

[0475] "Means of analysis using natural language processing" refers to the process of analyzing information using natural language processing techniques such as morphological analysis and sentiment analysis.

[0476] "Means for detecting risk factors and calculating a risk score" refers to the function of identifying elements that indicate suicide risk from the analysis results and calculating a comprehensive assessment of this as a score.

[0477] "Means for providing warnings" refers to a function that notifies the user of warnings and support information when a risk factor is detected.

[0478] The "means for receiving feedback" refers to a method for collecting feedback from the user as to whether the assistance provided was appropriate.

[0479] "Means of learning and improving the system" refers to the ability to retrain AI models based on feedback and improve the overall performance and accuracy of the system.

[0480] This invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. The system collects various information from users and provides alerts based on the results of analyzing that information. It also collects feedback from users and learns and improves the system's accuracy.

[0481] How we collect your information

[0482] Users download the app, create a new account, and the app prompts them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this information and sends it to a server.

[0483] Information preprocessing

[0484] The server stores the received information in temporary storage. Then, the server filters out incomplete data and noise, tokenizes the text information, and uses a natural language processing (NLP) library (e.g., SpaCy or NLTK) to perform morphological analysis of the information and extract useful information.

[0485] Risk factor detection

[0486] The server inputs the preprocessed information into natural language processing models (e.g., BERT or GPT-3). These models analyze keywords and sentiment to detect suicide risk factors. They also analyze sensor data to detect anomalous patterns. From this, the server generates an overall risk score and generates an alert if the score exceeds a certain threshold.

[0487] Providing alerts

[0488] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[0489] Gathering feedback and training the system

[0490] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[0491] Specific examples

[0492] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[0493] Example prompts to input to a generative AI model:

[0494] User 01 sends the following message: "I've been feeling down lately." Conduct a sentiment analysis of this message and assess its risk factors.

[0495] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0496] Step 1:

[0497] A user downloads the app and creates a new account.

[0498] Input: Account information entered by the user (name, email address, password)

[0499] Specific operation: After downloading the application from the app store, the user installs it, then launches the app, enters the required information on the new account creation screen, and presses the "Register" button.

[0500] Output: The account information is sent to the server and the new account is saved in the database.

[0501] Step 2:

[0502] Users consent to providing information and text messages, social media posts, and sensor data are collected.

[0503] Input: User-entered consent to provide information

[0504] Specific operation: The user consents to providing information on the account settings screen. After consent, the device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server.

[0505] Output: The collected information is sent to the server and stored in temporary storage.

[0506] Step 3:

[0507] The server preprocesses the received information.

[0508] Input: User text messages, social media posts, and sensor data sent from the device

[0509] How it works: The server stores the collected information in temporary storage, filters out incomplete data and noise, and then uses a natural language processing library (e.g., SpaCy, NLTK) to tokenize the text information and perform morphological analysis.

[0510] Output: Preprocessed and clean information

[0511] Step 4:

[0512] The server inputs the preprocessed information into a natural language processing model to detect risk factors.

[0513] Input: Preprocessed information

[0514] How it works: The server inputs the preprocessed information into a generative AI model (e.g., GPT-3) to perform sentiment analysis and keyword extraction. It then analyzes the sensor data to detect anomalous patterns. This allows the server to calculate a risk score for each data point.

[0515] Output: Risk factor detection results and risk score

[0516] Step 5:

[0517] If the risk score exceeds a certain threshold, the server generates an alert and sends it to the terminal.

[0518] Input: Risk score

[0519] Specific operation: The server evaluates whether the risk score exceeds the set threshold. If it does, it generates an alert message and sends it to the terminal.

[0520] Output: Alert information

[0521] Step 6:

[0522] The device notifies the user of the alert.

[0523] Input: Alert information sent from the server

[0524] What it does: The device immediately notifies the user of an alert message, which includes a warning message, a chat link with a counselor, and emergency contact information.

[0525] Output: The alert message that is displayed to the user.

[0526] Step 7:

[0527] The user provides feedback and the feedback information is sent to the server.

[0528] Input: User feedback information (e.g., appropriateness of assistance)

[0529] Specific operation: The user provides feedback within the app on whether the provided assistance was appropriate. The device then sends this feedback information to the server.

[0530] Output: Feedback information sent to the server

[0531] Step 8:

[0532] The server uses feedback to learn and improve the system.

[0533] Input: User feedback information

[0534] Specific operation: The server uses the received feedback information to retrain the AI ​​model and improve the accuracy of the entire system, thereby improving the accuracy of the next risk factor detection and alert.

[0535] Output: Improved AI models and increased system accuracy

[0536] (Application example 1)

[0537] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0538] In modern society, mental health issues can increase the risk of suicide, creating a need for early detection and rapid response. However, existing systems lack sufficient means for integrating emotion analysis and biometric sensor data to assess risk, making it difficult to provide support at the appropriate time. The present invention aims to solve this problem by more accurately monitoring a user's mental health status and providing support at the appropriate time.

[0539] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0540] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for detecting risk factors from the preprocessed data, means for providing an alert to the user based on the detected risk factors, means for receiving feedback from the user, means for learning and improving the system based on the feedback, means for evaluating the user's emotions using an emotion analysis model, and means for analyzing biosensor data to detect abnormal patterns. This makes it possible to comprehensively analyze a variety of user data, detect risks early, and provide support at an appropriate time.

[0541] "Means for collecting data from users" refers to devices or software systems for obtaining users' text messages, social media posts, sensor data, etc.

[0542] "Means for pre-processing collected data" refers to processes or algorithms used to organize, filter, tokenize, etc. the data obtained.

[0543] "Risk factor detection tools" are algorithms or models that analyze pre-processed data and assess suicide risk based on specific patterns or keywords.

[0544] An "alert providing means" is a device or system for sending a warning or caution to a user based on a detected risk factor.

[0545] The "means for receiving feedback from users" refers to an interface or data collection system for collecting ratings and comments provided by users.

[0546] "Means of learning and improving the system" refers to the process of updating AI models and algorithms based on collected feedback to improve their accuracy.

[0547] The "means for evaluating emotions using an emotion analysis model" is a system that extracts and evaluates emotions from text data using a natural language processing model or other analysis tools.

[0548] "Means for analyzing biometric sensor data to detect abnormal patterns" refers to algorithms and analytical tools that analyze biometric data such as heart rate and sleep patterns to detect patterns that are out of the ordinary.

[0549] The present invention relates to a system for monitoring a user's mental health status, detecting risks early, and providing support at an appropriate time. The system is composed of the following main components:

[0550] 1. Data collection methods:

[0551] It collects users' text messages, social media posts, and biometric sensor data (e.g., heart rate, sleep patterns) using devices such as smartphones, smart glasses, and smart watches.

[0552] 2. Data preprocessing methods:

[0553] The server stores the collected data in temporary storage. It then filters out incomplete data and noise and tokenizes the text data. It uses a natural language processing library (e.g., TextBlob) to perform morphological analysis of the data and extract useful information. It also filters and cleans the sensor data in the same way.

[0554] 3. Risk factor detection methods:

[0555] The preprocessed data is input into a natural language processing model (e.g., an NLP model). This model analyzes emotions from the text and calculates an emotion score. It also analyzes biometric sensor data to detect abnormal patterns (e.g., high heart rate or abnormal sleep patterns). The server comprehensively evaluates this information and identifies risk factors for suicide.

[0556] 4. Alert Delivery Method:

[0557] If a risk factor is detected, the server will send an alert to the user's device, which will include a warning message, a chat link with a counselor, and emergency contact information, allowing the user to receive immediate and appropriate assistance.

[0558] 5. Feedback collection methods:

[0559] After providing the assistance, the user provides feedback on whether the assistance provided was appropriate, and the terminal transmits this feedback information to the server.

[0560] 6. System Learning and Improvement Measures:

[0561] The server retrains the AI ​​model based on feedback collected from users to improve the accuracy of the system.

[0562] Example

[0563] For example, if User A sends a message saying, "I've been feeling down lately," this message is collected from a smartphone app. The server tokenizes this message and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to User A's smartphone saying, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[0564] Prompt Sentence Examples

[0565] When a user sends a message saying, "I've been feeling down lately," the emotion score detected from this message is 0.7, indicating a very negative state. The user's heart rate data also shows abnormalities, indicating that immediate assistance is required.

[0566] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0567] Step 1:

[0568] A user installs the application and creates a new account. The user's smartphone, smart glasses, or smartwatch becomes a device for collecting text messages, social media posts, and biometric sensor data. This collects user input data (e.g., a text message saying "I've been feeling down lately") and sensor data (e.g., heart rate, sleep patterns).

[0569] Input: User text messages, social media posts, biometric sensor data

[0570] Output: Raw data sent to the server

[0571] Step 2:

[0572] The server stores the collected data in temporary storage. Next, it performs preprocessing, filtering out incomplete data and noise and tokenizing the text data. Specifically, it uses a natural language processing library (e.g., TextBlob) to perform morphological analysis and extract useful information. It also cleans the sensor data in a similar way, removing incomplete values.

[0573] Input: Collected raw data

[0574] Output: Filtered and tokenized preprocessed data

[0575] Step 3:

[0576] The server inputs the preprocessed text data into a natural language processing model (NLP model). The NLP model performs sentiment analysis and calculates a sentiment score for the text (e.g., -0.7). At the same time, it also detects abnormal patterns in the sensor data (e.g., a heart rate above 100). This allows for a comprehensive calculation of suicide risk.

[0577] Input: Preprocessed text data and sensor data

[0578] Output: Sentiment score and abnormal pattern detection results

[0579] Step 4:

[0580] The server evaluates risk factors based on the emotion score and abnormal patterns in the sensor data. If the risk factor is determined to be high (e.g., emotion score -0.7 or heart rate above 100), an alert is generated. The generated alert includes a warning message, a link to counseling services, and emergency contact information.

[0581] Input: Sentiment scores and anomaly pattern detection results

[0582] Output: Generated alerts

[0583] Step 5:

[0584] The server then sends the generated alert to the user's device. The user immediately receives the alert on their smartphone or smart glasses and checks the notification. The notification includes a warning such as, "Your recent messages have caused you concern. Counseling services are available," along with emergency contact information.

[0585] Input: Generated alert

[0586] Output: Notification displayed on the user's device

[0587] Step 6:

[0588] The user provides feedback on whether the provided assistance was appropriate. The terminal sends this feedback information to the server, which uses it to improve the system.

[0589] Input: User feedback

[0590] Output: Feedback information sent to the server

[0591] Step 7:

[0592] The server uses the collected feedback information to retrain the AI ​​model and improve the accuracy of the system, which will improve the accuracy and speed of future risk detection.

[0593] Input: Feedback information

[0594] Output: Retrained AI model

[0595] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0596] MODE FOR CARRYING OUT THE INVENTION

[0597] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects a variety of data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the system's accuracy. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it enables more accurate detection of risk factors. Detailed embodiments of the system are described below.

[0598] How we collect your data

[0599] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[0600] Data Preprocessing

[0601] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[0602] Filter out incomplete data and noise.

[0603] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library to extract useful information.

[0604] Emotion recognition by emotion engine

[0605] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[0606] Categorize emotions as negative, positive, or neutral.

[0607] A score is generated for each emotion category.

[0608] Risk factor detection

[0609] The server comprehensively analyzes the emotion scores output from the emotion engine and the sensor data. Specifically, it performs the following steps:

[0610] Keyword analysis: Detect specific risky keywords.

[0611] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[0612] Behavioral pattern detection: Analyze sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[0613] Calculating the overall risk score

[0614] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[0615] Providing alerts

[0616] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[0617] A message expressing concern

[0618] Chat link with a counselor

[0619] Emergency contact information

[0620] Gathering feedback and training the system

[0621] The user checks the notification and enters feedback into the device on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and enables it to provide more accurate alerts and assistance in the future.

[0622] Specific examples

[0623] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[0624] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[0625] The processing flow will be explained below.

[0626] Step 1:

[0627] Users download the app and create a new account by entering basic information (such as name, age, and gender). They then grant permission for access to text messages, social media posts, and sensor data.

[0628] Step 2:

[0629] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[0630] Step 3:

[0631] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[0632] Step 4:

[0633] The server stores the received data in temporary storage. Then, data preprocessing begins. Specifically, it performs the following processes:

[0634] Filter out incomplete data and noise.

[0635] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[0636] Step 5:

[0637] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[0638] Classify the sentiment of each text message or post as positive, negative, or neutral.

[0639] A score is generated for each emotion category.

[0640] Step 6:

[0641] The server uses natural language processing models to analyze the risk factors in the text data. Specifically, it performs the following tasks:

[0642] Keyword analysis: Detect specific risky keywords.

[0643] Sentiment Analysis: Evaluating the sentiment of text data and integrating the scores generated by the sentiment engine.

[0644] Step 7:

[0645] The server integrates the emotion engine score and the results of risk factor analysis of the text data to calculate an overall risk score, and generates an emergency alert if the score exceeds a certain threshold.

[0646] Step 8:

[0647] The server analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Examples include:

[0648] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[0649] Decreased activity level

[0650] Step 9:

[0651] The server integrates the emotion score, the text data risk score, and the sensor data results to make a final risk assessment. If the overall risk score exceeds a threshold, an alert is generated.

[0652] Step 10:

[0653] The device will notify the user of the alert information received from the server. The notification will be displayed as a popup and will include the following:

[0654] A message expressing concern

[0655] Chat link with a counselor

[0656] Emergency contact information

[0657] Step 11:

[0658] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[0659] Step 12:

[0660] The user inputs feedback into the terminal, including an evaluation of the timing and content of the alert, on whether the alert and the provided assistance were appropriate.

[0661] Step 13:

[0662] The device sends feedback to the server, which receives it and uses it to retrain the AI ​​model, improving the system's accuracy and enabling it to provide better alerts and assistance in the future.

[0663] This is the specific processing flow of this system, which combines an emotion engine, making it possible to grasp the user's emotional state more accurately and provide support at the appropriate time.

[0664] Example 2

[0665] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0666] In recent years, mental health issues such as psychological stress and depression have become a major social issue. In particular, it is important to detect suicide risk factors early and provide support at the appropriate time. However, conventional systems can sometimes have difficulty accurately understanding a user's mental state and intervening quickly. For this reason, there is a demand for a system that can analyze a variety of data from users and detect risk factors with high accuracy.

[0667] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0668] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for inputting the preprocessed data into an emotion analysis engine to generate an emotion score, and means for analyzing the emotion score and sensor data to detect risk factors, thereby enabling a highly accurate analysis of the user's mental state and early detection of risk factors.

[0669] "User" refers to any individual or entity that uses the System.

[0670] "Data" refers to information collected from users, such as text messages, social media posts, and sensor data.

[0671] "Means for collection" refers to hardware and software for collecting data with the user's permission.

[0672] "Preprocessing means" refers to the process by which collected data is filtered, tokenized, and prepared for analysis.

[0673] "Sentiment analysis engine" refers to an algorithm or model for generating sentiment scores based on pre-processed text data.

[0674] "Emotion score" refers to a numerical representation of the intensity and type of emotion extracted from text data.

[0675] "Sensor data" refers to data collected from devices that measure a user's activity patterns and biometric information.

[0676] "Risk factors" refer to certain behaviors or psychological states that indicate a user is at increased risk of suicide.

[0677] An "alert" refers to a message that notifies a user or interested party that a risk factor has been detected.

[0678] "Feedback" refers to the user providing the system with their evaluation and opinion of the assistance provided.

[0679] "Learn and improve" refers to the process of retraining the system's algorithms and models based on feedback to improve performance.

[0680] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[0681] How we collect your data

[0682] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[0683] Data Preprocessing

[0684] The server stores the received data in temporary storage, after which it performs preprocessing on the data.

[0685] Specifically, it filters out incomplete data and noise, tokenizes the text data, and performs morphological analysis using natural language processing (NLP) libraries (e.g., NLTK and spaCy) to extract useful information.

[0686] Emotion recognition by emotion engine

[0687] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google Cloud Natural Language API), which classifies sentiment into negative, positive, and neutral and generates a score for each sentiment category.

[0688] Risk factor detection

[0689] The server comprehensively analyzes the emotion score output from the emotion engine and the sensor data. Specifically, it detects specific risk keywords, evaluates the emotion of the text data, and generates a score: positive, negative, or neutral. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[0690] Calculating the overall risk score

[0691] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score. If this overall risk score exceeds a certain threshold, the server generates an emergency alert.

[0692] Providing alerts

[0693] The device will notify the user of the alert received from the server. The notification will be displayed as a popup and will include the following:

[0694] A message expressing concern

[0695] Chat link with a counselor

[0696] Emergency contact information

[0697] Gathering feedback and training the system

[0698] The user checks the notification and enters feedback on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and allows it to provide more accurate alerts and assistance in the future.

[0699] Specific examples

[0700] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the result is determined to have a high negative score, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[0701] This allows the system to detect the user's mental state early and provide support at the appropriate time.

[0702] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0703] Step 1:

[0704] Users download the app and create a new account, entering basic information such as name, age, and gender, and granting access to text messages, social media posts, and sensor data, which is then sent by the device to a server.

[0705] Input: Name, age, gender, text messages, social media posts, sensor data

[0706] Output: Send basic information and permission information to the server

[0707] Step 2:

[0708] The server receives the user's basic information and permission information and stores it in temporary storage. The device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server. The server stores the received data in temporary storage.

[0709] Input: User basic information, text messages, social media posts, sensor data

[0710] Output: Saving data to temporary storage

[0711] Step 3:

[0712] The server begins preprocessing the collected data by filtering out incomplete data and noise, tokenizing the text data, and extracting useful information from the data by performing morphological analysis using a natural language processing (NLP) library (e.g., NLTK or spaCy).

[0713] Input: Raw data stored in temporary storage

[0714] Output: Preprocessed data (filtered, tokenized, morphologically analyzed)

[0715] Step 4:

[0716] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google Cloud Natural Language API), which generates a sentiment score for the text and classifies it into negative, positive, or neutral categories.

[0717] Input: Preprocessed text data

[0718] Output: Sentiment scores (negative, positive, neutral scores)

[0719] Step 5:

[0720] The server comprehensively analyzes the emotion score and sensor data. Specifically, it detects specific risk keywords and evaluates the emotion score. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Based on the results of this analysis, it generates individual scores.

[0721] Input: emotion scores, sensor data

[0722] Output: Risk factor score

[0723] Step 6:

[0724] The server combines the generated individual scores to calculate an overall risk score, and if this overall risk score exceeds a certain threshold, the server generates an emergency alert and sends it to the device.

[0725] Input: Risk factor score

[0726] Output: Overall risk score, emergency alert

[0727] Step 7:

[0728] The device notifies the user of emergency alerts received from the server. The notification appears as a pop-up and includes a message of concern, a chat link to a counselor, and emergency contact information.

[0729] Input: Emergency Alert

[0730] Output: A popup notification to the user

[0731] Step 8:

[0732] The user checks the notification and inputs feedback on whether the provided assistance was appropriate or not into the terminal, which then transmits this feedback information to the server.

[0733] Input: Feedback

[0734] Output: Sending feedback to the server

[0735] Step 9:

[0736] The server uses the received feedback data to retrain the AI ​​model, improving the system's performance and enabling it to provide more accurate alerts and assistance in the future.

[0737] Input: Feedback data

[0738] Output: Retrained AI model

[0739] (Application example 2)

[0740] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0741] In modern society, it is increasingly important to maintain users' mental health, detect risk factors early, and provide appropriate support. However, conventional systems have had difficulty properly collecting and analyzing diverse data and detecting risk factors in real time. Furthermore, a lack of means to provide support at the appropriate time has made it difficult to intervene effectively to ensure users' mental health. Therefore, the present invention aims to solve these problems by detecting users' mental states early and providing appropriate support.

[0742] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0743] In this invention, the server includes means for collecting data from the user, means for preprocessing the collected data, and means for detecting risk factors from the preprocessed data. This allows for the use of emotion scores and behavioral patterns in detecting risk factors. Furthermore, by using a means for providing notifications via a smartphone or smart glasses, it is possible to provide prompt and accurate alerts to the user. Furthermore, the system can receive feedback from the user and continue to learn and improve based on that feedback, thereby enabling continuous monitoring of the user's mental health and providing appropriate support.

[0744] "User" means an individual or organization that uses this system.

[0745] A "means for collecting data" is a device or function for transmitting user text messages, social media posts, sensor data, etc. to a server.

[0746] A "preprocessing means" is a device or function for organizing collected data, filtering out incomplete data and noise, and converting it into an analyzable format.

[0747] The "means for detecting risk factors" is a device or function for evaluating the user's mental health status and behavioral patterns from the preprocessed data and recognizing abnormal conditions and risks.

[0748] A "means for providing an alert" is a device or function for providing a warning or counseling suggestion to a user when a risk factor is detected.

[0749] A "means for receiving feedback" is a device or function for informing the server how the user responded to the alert or whether the assistance provided was adequate.

[0750] A "means for learning and improving the system" is a device or function that updates the system's algorithms based on user feedback to improve accuracy and effectiveness.

[0751] An "emotion score" is a numerical assessment of a user's emotional state based on text data and other data.

[0752] "Behavioral patterns" refer to a user's daily behavior and lifestyle habits analyzed from sensor data.

[0753] A "smartphone" is a portable communication device that has functions such as voice calls, text message sending, and internet connection.

[0754] "Smart glasses" are eyeglass-type devices incorporating display devices and sensors, and are devices that provide visual information to users.

[0755] A system for implementing this invention mainly uses the following hardware and software. The hardware uses a smartphone or smart glasses. The software includes a natural language processing (NLP) library, an emotion recognition engine, and cloud data storage. For example, "spaCy" or "NLTK" can be used as the natural language processing library. Possible emotion recognition engines include "Amazon Comprehend" or "Google Cloud Natural Language API." Furthermore, "AWS S3" or "Google Cloud Storage" is used as the cloud data storage. "React Native" or "Flutter" is used for mobile app development.

[0756] The server collects user data such as text messages, social media posts, and sensor data. During data preprocessing, the collected data is organized and incomplete data and noise are filtered. The preprocessed data is input into a natural language processing model and an emotion recognition engine. The server uses emotion scores and behavioral patterns to detect risk factors and provide alerts to users.

[0757] As a concrete example, a user may send a message via a smartphone app saying, "I'm feeling depressed." The server tokenizes the message and analyzes it using a natural language processing model and an emotion recognition engine. If the text is determined to indicate a high risk factor, the server will provide an alert to the user via their smartphone or smart glasses. The alert message may say, "Your recent messages have caused us concern. Counseling services are available."

[0758] The prompt statement in the example program is:

[0759] Input prompt: User types "I'm feeling down"

[0760] System Output: "Your recent messages have me concerned. Counseling services are available."

[0761] This allows users to receive appropriate assistance quickly. The system also learns and improves based on user feedback, which will enable more accurate risk factor detection and assistance provision in the future.

[0762] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0763] Step 1:

[0764] A user uses a smartphone or smart glasses to input a text message, for example, the user inputs the message "I'm feeling down," which is sent to the system.

[0765] Step 2:

[0766] The terminal sends the entered text message to the server, where it is temporarily stored.

[0767] Step 3:

[0768] The server preprocesses the received text message. The input is the text message, and the output is tokenized text. Specifically, it uses a natural language processing library (e.g., spaCy) to tokenize the message and perform morphological analysis.

[0769] Step 4:

[0770] The server inputs the preprocessed text data into an emotion recognition engine. The input is tokenized text, and the output is an emotion score. This process uses services such as Amazon Comprehend or Google Cloud Natural Language API to generate an emotion score (e.g., negative, positive, neutral).

[0771] Step 5:

[0772] The server combines the generated emotion scores with the sensor data to calculate an overall risk score. Specifically, it numerically evaluates and combines the emotion scores and behavioral pattern data (e.g., sleep patterns, activity levels). This results in an overall risk score.

[0773] Step 6:

[0774] If the overall risk score exceeds a certain threshold, the server generates an alert and notifies the device. Specifically, the notification content is created as a message such as "Your recent messages have caused you concern. Counseling services are available." and sent to the smartphone or smart glasses.

[0775] Step 7:

[0776] The device displays the received alert notification to the user, who can then check the alert and access information such as counseling services or emergency contacts as needed.

[0777] Step 8:

[0778] The user inputs feedback about the alert into the terminal, such as "The notified support was appropriate" or "I used the service."

[0779] Step 9:

[0780] The terminal transmits the feedback from the user to the server, which stores the feedback.

[0781] Step 10:

[0782] The server uses the feedback it receives to learn and improve the system. Specifically, it inputs the feedback data into a generative AI model, retraining the algorithm and improving the overall system performance, which will enable more accurate risk factor detection and assistance provision in the future.

[0783] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0784] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0785] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0786] [Third embodiment]

[0787] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0788] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0789] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0790] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0791] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0792] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0793] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0794] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0795] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0796] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0797] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0798] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0799] MODE FOR CARRYING OUT THE INVENTION

[0800] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[0801] How we collect your data

[0802] Users download the app, create a new account, and then the app asks them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this data and sends it to a server.

[0803] Data Preprocessing

[0804] The server stores the received data in temporary storage. Then, it filters out incomplete data and noise, tokenizes the text data, and uses natural language processing (NLP) libraries to perform morphological analysis of the data and extract useful information.

[0805] Risk factor detection

[0806] The server inputs the preprocessed data into a natural language processing model. This model performs keyword and sentiment analysis to detect suicide risk factors. It also analyzes the sensor data to detect abnormal patterns. From this, the server scores the overall risk and generates an alert if the score exceeds a certain threshold.

[0807] Providing alerts

[0808] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[0809] Gathering feedback and training the system

[0810] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[0811] Specific examples

[0812] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[0813] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[0814] The processing flow will be explained below.

[0815] Step 1:

[0816] Users download the app and create a new account, entering basic information like name, age, and gender, and granting access to text messages, social media posts, and sensor data.

[0817] Step 2:

[0818] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[0819] Step 3:

[0820] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[0821] Step 4:

[0822] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[0823] Filter out incomplete data and noise.

[0824] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[0825] Step 5:

[0826] The server inputs the preprocessed data into a natural language processing model, which performs the following analysis:

[0827] Keyword analysis: Detect specific risky keywords.

[0828] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[0829] Step 6:

[0830] The server analyzes the sensor data to detect abnormal behavioral patterns and lifestyle changes, including:

[0831] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[0832] Decreased activity level

[0833] Step 7:

[0834] The server combines the scores generated from the text data and sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[0835] Step 8:

[0836] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[0837] A message expressing concern

[0838] Chat link with a counselor

[0839] Emergency contact information

[0840] Step 9:

[0841] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[0842] Step 10:

[0843] The user inputs feedback into the terminal on whether the alert and the provided assistance were appropriate, including an evaluation of the timing and content of the alert.

[0844] Step 11:

[0845] The device sends the collected feedback to a server, which receives it and uses the feedback data to retrain the AI ​​model, improving the system's performance and providing more accurate alerts and assistance in the future.

[0846] As described above, this system collects and analyzes user data and provides support at the appropriate time, thereby detecting risk factors for suicide early and helping to save lives.

[0847] Example 1

[0848] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0849] Suicide prevention is an important issue in modern society. However, conventional methods currently struggle to detect a user's psychological state early and provide appropriate support in a timely manner. In particular, it is technically difficult to detect psychological risk factors early from personal text messages, social media posts, and daily activity data. Therefore, there is a need for a system that can monitor a user's mental health state in real time and provide necessary support promptly.

[0850] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0851] In this invention, the server includes means for collecting information from the user, means for temporarily storing the information, means for cleansing and tokenizing the information, means for analyzing the information using natural language processing, means for detecting risk factors and calculating a risk score, means for providing a warning, means for receiving feedback, and means for learning and improving the system, thereby enabling early detection of the user's psychological state and providing appropriate support in a timely manner.

[0852] "User" refers to an individual who uses this system.

[0853] "Information" refers to data collected from users, including text messages, social media posts, and activity data.

[0854] "Means of collection" refers to the functionality of an application or device to obtain information from a user.

[0855] "Temporary storage means" refers to a storage system or database for temporarily storing collected information.

[0856] "Cleansing and tokenization methods" refers to technologies that remove incomplete data and noise from information and perform processing to segment and identify text data.

[0857] "Means of analysis using natural language processing" refers to the process of analyzing information using natural language processing techniques such as morphological analysis and sentiment analysis.

[0858] "Means for detecting risk factors and calculating a risk score" refers to the function of identifying elements that indicate suicide risk from the analysis results and calculating a comprehensive assessment of this as a score.

[0859] "Means for providing warnings" refers to a function that notifies the user of warnings and support information when a risk factor is detected.

[0860] The "means for receiving feedback" refers to a method for collecting feedback from the user as to whether the assistance provided was appropriate.

[0861] "Means of learning and improving the system" refers to the ability to retrain AI models based on feedback and improve the overall performance and accuracy of the system.

[0862] This invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. The system collects various information from users and provides alerts based on the results of analyzing that information. It also collects feedback from users and learns and improves the system's accuracy.

[0863] How we collect your information

[0864] Users download the app, create a new account, and the app prompts them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this information and sends it to a server.

[0865] Information preprocessing

[0866] The server stores the received information in temporary storage. Then, the server filters out incomplete data and noise, tokenizes the text information, and uses a natural language processing (NLP) library (e.g., SpaCy or NLTK) to perform morphological analysis of the information and extract useful information.

[0867] Risk factor detection

[0868] The server inputs the preprocessed information into natural language processing models (e.g., BERT or GPT-3). These models analyze keywords and sentiment to detect suicide risk factors. They also analyze sensor data to detect anomalous patterns. From this, the server generates an overall risk score and generates an alert if the score exceeds a certain threshold.

[0869] Providing alerts

[0870] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[0871] Gathering feedback and training the system

[0872] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[0873] Specific examples

[0874] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[0875] Example prompts to input to a generative AI model:

[0876] User 01 sends the following message: "I've been feeling down lately." Conduct a sentiment analysis of this message and assess its risk factors.

[0877] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0878] Step 1:

[0879] A user downloads the app and creates a new account.

[0880] Input: Account information entered by the user (name, email address, password)

[0881] Specific operation: After downloading the application from the app store, the user installs it, then launches the app, enters the required information on the new account creation screen, and presses the "Register" button.

[0882] Output: The account information is sent to the server and the new account is saved in the database.

[0883] Step 2:

[0884] Users consent to providing information and text messages, social media posts, and sensor data are collected.

[0885] Input: User-entered consent to provide information

[0886] Specific operation: The user consents to providing information on the account settings screen. After consent, the device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server.

[0887] Output: The collected information is sent to the server and stored in temporary storage.

[0888] Step 3:

[0889] The server preprocesses the received information.

[0890] Input: User text messages, social media posts, and sensor data sent from the device

[0891] How it works: The server stores the collected information in temporary storage, filters out incomplete data and noise, and then uses a natural language processing library (e.g., SpaCy, NLTK) to tokenize the text information and perform morphological analysis.

[0892] Output: Preprocessed and clean information

[0893] Step 4:

[0894] The server inputs the preprocessed information into a natural language processing model to detect risk factors.

[0895] Input: Preprocessed information

[0896] How it works: The server inputs the preprocessed information into a generative AI model (e.g., GPT-3) to perform sentiment analysis and keyword extraction. It then analyzes the sensor data to detect anomalous patterns. This allows the server to calculate a risk score for each data point.

[0897] Output: Risk factor detection results and risk score

[0898] Step 5:

[0899] If the risk score exceeds a certain threshold, the server generates an alert and sends it to the terminal.

[0900] Input: Risk score

[0901] Specific operation: The server evaluates whether the risk score exceeds the set threshold. If it does, it generates an alert message and sends it to the terminal.

[0902] Output: Alert information

[0903] Step 6:

[0904] The device notifies the user of the alert.

[0905] Input: Alert information sent from the server

[0906] What it does: The device immediately notifies the user of an alert message, which includes a warning message, a chat link with a counselor, and emergency contact information.

[0907] Output: The alert message that is displayed to the user.

[0908] Step 7:

[0909] The user provides feedback and the feedback information is sent to the server.

[0910] Input: User feedback information (e.g., appropriateness of assistance)

[0911] Specific operation: The user provides feedback within the app on whether the provided assistance was appropriate. The device then sends this feedback information to the server.

[0912] Output: Feedback information sent to the server

[0913] Step 8:

[0914] The server uses feedback to learn and improve the system.

[0915] Input: User feedback information

[0916] Specific operation: The server uses the received feedback information to retrain the AI ​​model and improve the accuracy of the entire system, thereby improving the accuracy of the next risk factor detection and alert.

[0917] Output: Improved AI models and increased system accuracy

[0918] (Application example 1)

[0919] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0920] In modern society, mental health issues can increase the risk of suicide, creating a need for early detection and rapid response. However, existing systems lack sufficient means for integrating emotion analysis and biometric sensor data to assess risk, making it difficult to provide support at the appropriate time. The present invention aims to solve this problem by more accurately monitoring a user's mental health status and providing support at the appropriate time.

[0921] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0922] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for detecting risk factors from the preprocessed data, means for providing an alert to the user based on the detected risk factors, means for receiving feedback from the user, means for learning and improving the system based on the feedback, means for evaluating the user's emotions using an emotion analysis model, and means for analyzing biosensor data to detect abnormal patterns. This makes it possible to comprehensively analyze a variety of user data, detect risks early, and provide support at an appropriate time.

[0923] "Means for collecting data from users" refers to devices or software systems for obtaining users' text messages, social media posts, sensor data, etc.

[0924] "Means for pre-processing collected data" refers to processes or algorithms used to organize, filter, tokenize, etc. the data obtained.

[0925] "Risk factor detection tools" are algorithms or models that analyze pre-processed data and assess suicide risk based on specific patterns or keywords.

[0926] An "alert providing means" is a device or system for sending a warning or caution to a user based on a detected risk factor.

[0927] The "means for receiving feedback from users" refers to an interface or data collection system for collecting ratings and comments provided by users.

[0928] "Means of learning and improving the system" refers to the process of updating AI models and algorithms based on collected feedback to improve their accuracy.

[0929] The "means for evaluating emotions using an emotion analysis model" is a system that extracts and evaluates emotions from text data using a natural language processing model or other analysis tools.

[0930] "Means for analyzing biometric sensor data to detect abnormal patterns" refers to algorithms and analytical tools that analyze biometric data such as heart rate and sleep patterns to detect patterns that are out of the ordinary.

[0931] The present invention relates to a system for monitoring a user's mental health status, detecting risks early, and providing support at an appropriate time. The system is composed of the following main components:

[0932] 1. Data collection methods:

[0933] It collects users' text messages, social media posts, and biometric sensor data (e.g., heart rate, sleep patterns) using devices such as smartphones, smart glasses, and smart watches.

[0934] 2. Data preprocessing methods:

[0935] The server stores the collected data in temporary storage. It then filters out incomplete data and noise and tokenizes the text data. It uses a natural language processing library (e.g., TextBlob) to perform morphological analysis of the data and extract useful information. It also filters and cleans the sensor data in the same way.

[0936] 3. Risk factor detection methods:

[0937] The preprocessed data is input into a natural language processing model (e.g., an NLP model). This model analyzes emotions from the text and calculates an emotion score. It also analyzes biometric sensor data to detect abnormal patterns (e.g., high heart rate or abnormal sleep patterns). The server comprehensively evaluates this information and identifies risk factors for suicide.

[0938] 4. Alert Delivery Method:

[0939] If a risk factor is detected, the server will send an alert to the user's device, which will include a warning message, a chat link with a counselor, and emergency contact information, allowing the user to receive immediate and appropriate assistance.

[0940] 5. Feedback collection methods:

[0941] After providing the assistance, the user provides feedback on whether the assistance provided was appropriate, and the terminal transmits this feedback information to the server.

[0942] 6. System Learning and Improvement Measures:

[0943] The server retrains the AI ​​model based on feedback collected from users to improve the accuracy of the system.

[0944] Example

[0945] For example, if User A sends a message saying, "I've been feeling down lately," this message is collected from a smartphone app. The server tokenizes this message and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to User A's smartphone saying, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[0946] Prompt Sentence Examples

[0947] When a user sends a message saying, "I've been feeling down lately," the emotion score detected from this message is 0.7, indicating a very negative state. The user's heart rate data also shows abnormalities, indicating that immediate assistance is required.

[0948] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0949] Step 1:

[0950] A user installs the application and creates a new account. The user's smartphone, smart glasses, or smartwatch becomes a device for collecting text messages, social media posts, and biometric sensor data. This collects user input data (e.g., a text message saying "I've been feeling down lately") and sensor data (e.g., heart rate, sleep patterns).

[0951] Input: User text messages, social media posts, biometric sensor data

[0952] Output: Raw data sent to the server

[0953] Step 2:

[0954] The server stores the collected data in temporary storage. Next, it performs preprocessing, filtering out incomplete data and noise and tokenizing the text data. Specifically, it uses a natural language processing library (e.g., TextBlob) to perform morphological analysis and extract useful information. It also cleans the sensor data in a similar way, removing incomplete values.

[0955] Input: Collected raw data

[0956] Output: Filtered and tokenized preprocessed data

[0957] Step 3:

[0958] The server inputs the preprocessed text data into a natural language processing model (NLP model). The NLP model performs sentiment analysis and calculates a sentiment score for the text (e.g., -0.7). At the same time, it also detects abnormal patterns in the sensor data (e.g., a heart rate above 100). This allows for a comprehensive calculation of suicide risk.

[0959] Input: Preprocessed text data and sensor data

[0960] Output: Sentiment score and abnormal pattern detection results

[0961] Step 4:

[0962] The server evaluates risk factors based on the emotion score and abnormal patterns in the sensor data. If the risk factor is determined to be high (e.g., emotion score -0.7 or heart rate above 100), an alert is generated. The generated alert includes a warning message, a link to counseling services, and emergency contact information.

[0963] Input: Sentiment scores and anomaly pattern detection results

[0964] Output: Generated alerts

[0965] Step 5:

[0966] The server then sends the generated alert to the user's device. The user immediately receives the alert on their smartphone or smart glasses and checks the notification. The notification includes a warning such as, "Your recent messages have caused you concern. Counseling services are available," along with emergency contact information.

[0967] Input: Generated alert

[0968] Output: Notification displayed on the user's device

[0969] Step 6:

[0970] The user provides feedback on whether the provided assistance was appropriate. The terminal sends this feedback information to the server, which uses it to improve the system.

[0971] Input: User feedback

[0972] Output: Feedback information sent to the server

[0973] Step 7:

[0974] The server uses the collected feedback information to retrain the AI ​​model and improve the accuracy of the system, which will improve the accuracy and speed of future risk detection.

[0975] Input: Feedback information

[0976] Output: Retrained AI model

[0977] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0978] MODE FOR CARRYING OUT THE INVENTION

[0979] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects a variety of data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the system's accuracy. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it enables more accurate detection of risk factors. Detailed embodiments of the system are described below.

[0980] How we collect your data

[0981] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[0982] Data Preprocessing

[0983] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[0984] Filter out incomplete data and noise.

[0985] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library to extract useful information.

[0986] Emotion recognition by emotion engine

[0987] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[0988] Categorize emotions as negative, positive, or neutral.

[0989] A score is generated for each emotion category.

[0990] Risk factor detection

[0991] The server comprehensively analyzes the emotion scores output from the emotion engine and the sensor data. Specifically, it performs the following steps:

[0992] Keyword analysis: Detect specific risky keywords.

[0993] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[0994] Behavioral pattern detection: Analyze sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[0995] Calculating the overall risk score

[0996] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[0997] Providing alerts

[0998] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[0999] A message expressing concern

[1000] Chat link with a counselor

[1001] Emergency contact information

[1002] Gathering feedback and training the system

[1003] The user checks the notification and enters feedback into the device on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and enables it to provide more accurate alerts and assistance in the future.

[1004] Specific examples

[1005] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[1006] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[1007] The processing flow will be explained below.

[1008] Step 1:

[1009] Users download the app and create a new account by entering basic information (such as name, age, and gender). They then grant permission for access to text messages, social media posts, and sensor data.

[1010] Step 2:

[1011] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[1012] Step 3:

[1013] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[1014] Step 4:

[1015] The server stores the received data in temporary storage. Then, data preprocessing begins. Specifically, it performs the following processes:

[1016] Filter out incomplete data and noise.

[1017] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[1018] Step 5:

[1019] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[1020] Classify the sentiment of each text message or post as positive, negative, or neutral.

[1021] A score is generated for each emotion category.

[1022] Step 6:

[1023] The server uses natural language processing models to analyze the risk factors in the text data. Specifically, it performs the following tasks:

[1024] Keyword analysis: Detect specific risky keywords.

[1025] Sentiment Analysis: Evaluating the sentiment of text data and integrating the scores generated by the sentiment engine.

[1026] Step 7:

[1027] The server integrates the emotion engine score and the results of risk factor analysis of the text data to calculate an overall risk score, and generates an emergency alert if the score exceeds a certain threshold.

[1028] Step 8:

[1029] The server analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Examples include:

[1030] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[1031] Decreased activity level

[1032] Step 9:

[1033] The server integrates the emotion score, the text data risk score, and the sensor data results to make a final risk assessment. If the overall risk score exceeds a threshold, an alert is generated.

[1034] Step 10:

[1035] The device will notify the user of the alert information received from the server. The notification will be displayed as a popup and will include the following:

[1036] A message expressing concern

[1037] Chat link with a counselor

[1038] Emergency contact information

[1039] Step 11:

[1040] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[1041] Step 12:

[1042] The user inputs feedback into the terminal, including an evaluation of the timing and content of the alert, on whether the alert and the provided assistance were appropriate.

[1043] Step 13:

[1044] The device sends feedback to the server, which receives it and uses it to retrain the AI ​​model, improving the system's accuracy and enabling it to provide better alerts and assistance in the future.

[1045] This is the specific processing flow of this system, which combines an emotion engine, making it possible to grasp the user's emotional state more accurately and provide support at the appropriate time.

[1046] Example 2

[1047] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1048] In recent years, mental health issues such as psychological stress and depression have become a major social issue. In particular, it is important to detect suicide risk factors early and provide support at the appropriate time. However, conventional systems can sometimes have difficulty accurately understanding a user's mental state and intervening quickly. For this reason, there is a demand for a system that can analyze a variety of data from users and detect risk factors with high accuracy.

[1049] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1050] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for inputting the preprocessed data into an emotion analysis engine to generate an emotion score, and means for analyzing the emotion score and sensor data to detect risk factors, thereby enabling a highly accurate analysis of the user's mental state and early detection of risk factors.

[1051] "User" refers to any individual or entity that uses the System.

[1052] "Data" refers to information collected from users, such as text messages, social media posts, and sensor data.

[1053] "Means for collection" refers to hardware and software for collecting data with the user's permission.

[1054] "Preprocessing means" refers to the process by which collected data is filtered, tokenized, and prepared for analysis.

[1055] "Sentiment analysis engine" refers to an algorithm or model for generating sentiment scores based on pre-processed text data.

[1056] "Emotion score" refers to a numerical representation of the intensity and type of emotion extracted from text data.

[1057] "Sensor data" refers to data collected from devices that measure a user's activity patterns and biometric information.

[1058] "Risk factors" refer to certain behaviors or psychological states that indicate a user is at increased risk of suicide.

[1059] An "alert" refers to a message that notifies a user or interested party that a risk factor has been detected.

[1060] "Feedback" refers to the user providing the system with their evaluation and opinion of the assistance provided.

[1061] "Learn and improve" refers to the process of retraining the system's algorithms and models based on feedback to improve performance.

[1062] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[1063] How we collect your data

[1064] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[1065] Data Preprocessing

[1066] The server stores the received data in temporary storage, after which it performs preprocessing on the data.

[1067] Specifically, it filters out incomplete data and noise, tokenizes the text data, and performs morphological analysis using natural language processing (NLP) libraries (e.g., NLTK and spaCy) to extract useful information.

[1068] Emotion recognition by emotion engine

[1069] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google Cloud Natural Language API), which classifies sentiment into negative, positive, and neutral and generates a score for each sentiment category.

[1070] Risk factor detection

[1071] The server comprehensively analyzes the emotion score output from the emotion engine and the sensor data. Specifically, it detects specific risk keywords, evaluates the emotion of the text data, and generates a score: positive, negative, or neutral. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[1072] Calculating the overall risk score

[1073] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score. If this overall risk score exceeds a certain threshold, the server generates an emergency alert.

[1074] Providing alerts

[1075] The device will notify the user of the alert received from the server. The notification will be displayed as a popup and will include the following:

[1076] A message expressing concern

[1077] Chat link with a counselor

[1078] Emergency contact information

[1079] Gathering feedback and training the system

[1080] The user checks the notification and enters feedback on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and allows it to provide more accurate alerts and assistance in the future.

[1081] Specific examples

[1082] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the result is determined to have a high negative score, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[1083] This allows the system to detect the user's mental state early and provide support at the appropriate time.

[1084] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1085] Step 1:

[1086] Users download the app and create a new account, entering basic information such as name, age, and gender, and granting access to text messages, social media posts, and sensor data, which is then sent by the device to a server.

[1087] Input: Name, age, gender, text messages, social media posts, sensor data

[1088] Output: Send basic information and permission information to the server

[1089] Step 2:

[1090] The server receives the user's basic information and permission information and stores it in temporary storage. The device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server. The server stores the received data in temporary storage.

[1091] Input: User basic information, text messages, social media posts, sensor data

[1092] Output: Saving data to temporary storage

[1093] Step 3:

[1094] The server begins preprocessing the collected data by filtering out incomplete data and noise, tokenizing the text data, and extracting useful information from the data by performing morphological analysis using a natural language processing (NLP) library (e.g., NLTK or spaCy).

[1095] Input: Raw data stored in temporary storage

[1096] Output: Preprocessed data (filtered, tokenized, morphologically analyzed)

[1097] Step 4:

[1098] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google Cloud Natural Language API), which generates a sentiment score for the text and classifies it into negative, positive, or neutral categories.

[1099] Input: Preprocessed text data

[1100] Output: Sentiment scores (negative, positive, neutral scores)

[1101] Step 5:

[1102] The server comprehensively analyzes the emotion score and sensor data. Specifically, it detects specific risk keywords and evaluates the emotion score. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Based on the results of this analysis, it generates individual scores.

[1103] Input: emotion scores, sensor data

[1104] Output: Risk factor score

[1105] Step 6:

[1106] The server combines the generated individual scores to calculate an overall risk score, and if this overall risk score exceeds a certain threshold, the server generates an emergency alert and sends it to the device.

[1107] Input: Risk factor score

[1108] Output: Overall risk score, emergency alert

[1109] Step 7:

[1110] The device notifies the user of emergency alerts received from the server. The notification appears as a pop-up and includes a message of concern, a chat link to a counselor, and emergency contact information.

[1111] Input: Emergency Alert

[1112] Output: A popup notification to the user

[1113] Step 8:

[1114] The user checks the notification and inputs feedback on whether the provided assistance was appropriate or not into the terminal, which then transmits this feedback information to the server.

[1115] Input: Feedback

[1116] Output: Sending feedback to the server

[1117] Step 9:

[1118] The server uses the received feedback data to retrain the AI ​​model, improving the system's performance and enabling it to provide more accurate alerts and assistance in the future.

[1119] Input: Feedback data

[1120] Output: Retrained AI model

[1121] (Application example 2)

[1122] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1123] In modern society, it is increasingly important to maintain users' mental health, detect risk factors early, and provide appropriate support. However, conventional systems have had difficulty properly collecting and analyzing diverse data and detecting risk factors in real time. Furthermore, a lack of means to provide support at the appropriate time has made it difficult to intervene effectively to ensure users' mental health. Therefore, the present invention aims to solve these problems by detecting users' mental states early and providing appropriate support.

[1124] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1125] In this invention, the server includes means for collecting data from the user, means for preprocessing the collected data, and means for detecting risk factors from the preprocessed data. This allows for the use of emotion scores and behavioral patterns in detecting risk factors. Furthermore, by using a means for providing notifications via a smartphone or smart glasses, it is possible to provide prompt and accurate alerts to the user. Furthermore, the system can receive feedback from the user and continue to learn and improve based on that feedback, thereby enabling continuous monitoring of the user's mental health and providing appropriate support.

[1126] "User" means an individual or organization that uses this system.

[1127] A "means for collecting data" is a device or function for transmitting user text messages, social media posts, sensor data, etc. to a server.

[1128] A "preprocessing means" is a device or function for organizing collected data, filtering out incomplete data and noise, and converting it into an analyzable format.

[1129] The "means for detecting risk factors" is a device or function for evaluating the user's mental health status and behavioral patterns from the preprocessed data and recognizing abnormal conditions and risks.

[1130] A "means for providing an alert" is a device or function for providing a warning or counseling suggestion to a user when a risk factor is detected.

[1131] A "means for receiving feedback" is a device or function for informing the server how the user responded to the alert or whether the assistance provided was adequate.

[1132] A "means for learning and improving the system" is a device or function that updates the system's algorithms based on user feedback to improve accuracy and effectiveness.

[1133] An "emotion score" is a numerical assessment of a user's emotional state based on text data and other data.

[1134] "Behavioral patterns" refer to a user's daily behavior and lifestyle habits analyzed from sensor data.

[1135] A "smartphone" is a portable communication device that has functions such as voice calls, text message sending, and internet connection.

[1136] "Smart glasses" are eyeglass-type devices incorporating display devices and sensors, and are devices that provide visual information to users.

[1137] A system for implementing this invention mainly uses the following hardware and software. The hardware uses a smartphone or smart glasses. The software includes a natural language processing (NLP) library, an emotion recognition engine, and cloud data storage. For example, "spaCy" or "NLTK" can be used as the natural language processing library. Possible emotion recognition engines include "Amazon Comprehend" or "Google Cloud Natural Language API." Furthermore, "AWS S3" or "Google Cloud Storage" is used as the cloud data storage. "React Native" or "Flutter" is used for mobile app development.

[1138] The server collects user data such as text messages, social media posts, and sensor data. During data preprocessing, the collected data is organized and incomplete data and noise are filtered. The preprocessed data is input into a natural language processing model and an emotion recognition engine. The server uses emotion scores and behavioral patterns to detect risk factors and provide alerts to users.

[1139] As a concrete example, a user may send a message via a smartphone app saying, "I'm feeling depressed." The server tokenizes the message and analyzes it using a natural language processing model and an emotion recognition engine. If the text is determined to indicate a high risk factor, the server will provide an alert to the user via their smartphone or smart glasses. The alert message may say, "Your recent messages have caused us concern. Counseling services are available."

[1140] The prompt statement in the example program is:

[1141] Input prompt: User types "I'm feeling down"

[1142] System Output: "Your recent messages have me concerned. Counseling services are available."

[1143] This allows users to receive appropriate assistance quickly. The system also learns and improves based on user feedback, which will enable more accurate risk factor detection and assistance provision in the future.

[1144] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1145] Step 1:

[1146] A user uses a smartphone or smart glasses to input a text message, for example, the user inputs the message "I'm feeling down," which is sent to the system.

[1147] Step 2:

[1148] The terminal sends the entered text message to the server, where it is temporarily stored.

[1149] Step 3:

[1150] The server preprocesses the received text message. The input is the text message, and the output is tokenized text. Specifically, it uses a natural language processing library (e.g., spaCy) to tokenize the message and perform morphological analysis.

[1151] Step 4:

[1152] The server inputs the preprocessed text data into an emotion recognition engine. The input is tokenized text, and the output is an emotion score. This process uses services such as Amazon Comprehend or Google Cloud Natural Language API to generate an emotion score (e.g., negative, positive, neutral).

[1153] Step 5:

[1154] The server combines the generated emotion scores with the sensor data to calculate an overall risk score. Specifically, it numerically evaluates and combines the emotion scores and behavioral pattern data (e.g., sleep patterns, activity levels). This results in an overall risk score.

[1155] Step 6:

[1156] If the overall risk score exceeds a certain threshold, the server generates an alert and notifies the device. Specifically, the notification content is created as a message such as "Your recent messages have caused you concern. Counseling services are available." and sent to the smartphone or smart glasses.

[1157] Step 7:

[1158] The device displays the received alert notification to the user, who can then check the alert and access information such as counseling services or emergency contacts as needed.

[1159] Step 8:

[1160] The user inputs feedback about the alert into the terminal, such as "The notified support was appropriate" or "I used the service."

[1161] Step 9:

[1162] The terminal transmits the feedback from the user to the server, which stores the feedback.

[1163] Step 10:

[1164] The server uses the feedback it receives to learn and improve the system. Specifically, it inputs the feedback data into a generative AI model, retraining the algorithm and improving the overall system performance, which will enable more accurate risk factor detection and assistance provision in the future.

[1165] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1166] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1167] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1168] [Fourth embodiment]

[1169] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1170] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1171] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1172] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1173] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1174] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1175] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1176] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1177] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1178] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1179] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1180] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1181] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1182] MODE FOR CARRYING OUT THE INVENTION

[1183] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[1184] How we collect your data

[1185] Users download the app, create a new account, and then the app asks them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this data and sends it to a server.

[1186] Data Preprocessing

[1187] The server stores the received data in temporary storage. Then, it filters out incomplete data and noise, tokenizes the text data, and uses natural language processing (NLP) libraries to perform morphological analysis of the data and extract useful information.

[1188] Risk factor detection

[1189] The server inputs the preprocessed data into a natural language processing model. This model performs keyword and sentiment analysis to detect suicide risk factors. It also analyzes the sensor data to detect abnormal patterns. From this, the server scores the overall risk and generates an alert if the score exceeds a certain threshold.

[1190] Providing alerts

[1191] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[1192] Gathering feedback and training the system

[1193] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[1194] Specific examples

[1195] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[1196] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[1197] The processing flow will be explained below.

[1198] Step 1:

[1199] Users download the app and create a new account, entering basic information like name, age, and gender, and granting access to text messages, social media posts, and sensor data.

[1200] Step 2:

[1201] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[1202] Step 3:

[1203] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[1204] Step 4:

[1205] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[1206] Filter out incomplete data and noise.

[1207] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[1208] Step 5:

[1209] The server inputs the preprocessed data into a natural language processing model, which performs the following analysis:

[1210] Keyword analysis: Detect specific risky keywords.

[1211] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[1212] Step 6:

[1213] The server analyzes the sensor data to detect abnormal behavioral patterns and lifestyle changes, including:

[1214] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[1215] Decreased activity level

[1216] Step 7:

[1217] The server combines the scores generated from the text data and sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[1218] Step 8:

[1219] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[1220] A message expressing concern

[1221] Chat link with a counselor

[1222] Emergency contact information

[1223] Step 9:

[1224] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[1225] Step 10:

[1226] The user inputs feedback into the terminal on whether the alert and the provided assistance were appropriate, including an evaluation of the timing and content of the alert.

[1227] Step 11:

[1228] The device sends the collected feedback to a server, which receives it and uses the feedback data to retrain the AI ​​model, improving the system's performance and providing more accurate alerts and assistance in the future.

[1229] As described above, this system collects and analyzes user data and provides support at the appropriate time, thereby detecting risk factors for suicide early and helping to save lives.

[1230] Example 1

[1231] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1232] Suicide prevention is an important issue in modern society. However, conventional methods currently struggle to detect a user's psychological state early and provide appropriate support in a timely manner. In particular, it is technically difficult to detect psychological risk factors early from personal text messages, social media posts, and daily activity data. Therefore, there is a need for a system that can monitor a user's mental health state in real time and provide necessary support promptly.

[1233] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1234] In this invention, the server includes means for collecting information from the user, means for temporarily storing the information, means for cleansing and tokenizing the information, means for analyzing the information using natural language processing, means for detecting risk factors and calculating a risk score, means for providing a warning, means for receiving feedback, and means for learning and improving the system, thereby enabling early detection of the user's psychological state and providing appropriate support in a timely manner.

[1235] "User" refers to an individual who uses this system.

[1236] "Information" refers to data collected from users, including text messages, social media posts, and activity data.

[1237] "Means of collection" refers to the functionality of an application or device to obtain information from a user.

[1238] "Temporary storage means" refers to a storage system or database for temporarily storing collected information.

[1239] "Cleansing and tokenization methods" refers to technologies that remove incomplete data and noise from information and perform processing to segment and identify text data.

[1240] "Means of analysis using natural language processing" refers to the process of analyzing information using natural language processing techniques such as morphological analysis and sentiment analysis.

[1241] "Means for detecting risk factors and calculating a risk score" refers to the function of identifying elements that indicate suicide risk from the analysis results and calculating a comprehensive assessment of this as a score.

[1242] "Means for providing warnings" refers to a function that notifies the user of warnings and support information when a risk factor is detected.

[1243] The "means for receiving feedback" refers to a method for collecting feedback from the user as to whether the assistance provided was appropriate.

[1244] "Means of learning and improving the system" refers to the ability to retrain AI models based on feedback and improve the overall performance and accuracy of the system.

[1245] This invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. The system collects various information from users and provides alerts based on the results of analyzing that information. It also collects feedback from users and learns and improves the system's accuracy.

[1246] How we collect your information

[1247] Users download the app, create a new account, and the app prompts them to provide text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels). If the user agrees, the device periodically collects this information and sends it to a server.

[1248] Information preprocessing

[1249] The server stores the received information in temporary storage. Then, the server filters out incomplete data and noise, tokenizes the text information, and uses a natural language processing (NLP) library (e.g., SpaCy or NLTK) to perform morphological analysis of the information and extract useful information.

[1250] Risk factor detection

[1251] The server inputs the preprocessed information into natural language processing models (e.g., BERT or GPT-3). These models analyze keywords and sentiment to detect suicide risk factors. They also analyze sensor data to detect anomalous patterns. From this, the server generates an overall risk score and generates an alert if the score exceeds a certain threshold.

[1252] Providing alerts

[1253] When the device receives an alert from the server, it immediately notifies the user. The notification includes a warning message, a chat link to a counselor, and emergency contact information, so the user can receive appropriate support.

[1254] Gathering feedback and training the system

[1255] The user provides feedback on whether the assistance provided was appropriate, and the device sends this feedback information to the server, which uses it to retrain the AI ​​model and improve the system's accuracy.

[1256] Specific examples

[1257] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this point, User A can receive appropriate support.

[1258] Example prompts to input to a generative AI model:

[1259] User 01 sends the following message: "I've been feeling down lately." Conduct a sentiment analysis of this message and assess its risk factors.

[1260] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1261] Step 1:

[1262] A user downloads the app and creates a new account.

[1263] Input: Account information entered by the user (name, email address, password)

[1264] Specific operation: After downloading the application from the app store, the user installs it, then launches the app, enters the required information on the new account creation screen, and presses the "Register" button.

[1265] Output: The account information is sent to the server and the new account is saved in the database.

[1266] Step 2:

[1267] Users consent to providing information and text messages, social media posts, and sensor data are collected.

[1268] Input: User-entered consent to provide information

[1269] Specific operation: The user consents to providing information on the account settings screen. After consent, the device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server.

[1270] Output: The collected information is sent to the server and stored in temporary storage.

[1271] Step 3:

[1272] The server preprocesses the received information.

[1273] Input: User text messages, social media posts, and sensor data sent from the device

[1274] How it works: The server stores the collected information in temporary storage, filters out incomplete data and noise, and then uses a natural language processing library (e.g., SpaCy, NLTK) to tokenize the text information and perform morphological analysis.

[1275] Output: Preprocessed and clean information

[1276] Step 4:

[1277] The server inputs the preprocessed information into a natural language processing model to detect risk factors.

[1278] Input: Preprocessed information

[1279] How it works: The server inputs the preprocessed information into a generative AI model (e.g., GPT-3) to perform sentiment analysis and keyword extraction. It then analyzes the sensor data to detect anomalous patterns. This allows the server to calculate a risk score for each data point.

[1280] Output: Risk factor detection results and risk score

[1281] Step 5:

[1282] If the risk score exceeds a certain threshold, the server generates an alert and sends it to the terminal.

[1283] Input: Risk score

[1284] Specific operation: The server evaluates whether the risk score exceeds the set threshold. If it does, it generates an alert message and sends it to the terminal.

[1285] Output: Alert information

[1286] Step 6:

[1287] The device notifies the user of the alert.

[1288] Input: Alert information sent from the server

[1289] What it does: The device immediately notifies the user of an alert message, which includes a warning message, a chat link with a counselor, and emergency contact information.

[1290] Output: The alert message that is displayed to the user.

[1291] Step 7:

[1292] The user provides feedback and the feedback information is sent to the server.

[1293] Input: User feedback information (e.g., appropriateness of assistance)

[1294] Specific operation: The user provides feedback within the app on whether the provided assistance was appropriate. The device then sends this feedback information to the server.

[1295] Output: Feedback information sent to the server

[1296] Step 8:

[1297] The server uses feedback to learn and improve the system.

[1298] Input: User feedback information

[1299] Specific operation: The server uses the received feedback information to retrain the AI ​​model and improve the accuracy of the entire system, thereby improving the accuracy of the next risk factor detection and alert.

[1300] Output: Improved AI models and increased system accuracy

[1301] (Application example 1)

[1302] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1303] In modern society, mental health issues can increase the risk of suicide, creating a need for early detection and rapid response. However, existing systems lack sufficient means for integrating emotion analysis and biometric sensor data to assess risk, making it difficult to provide support at the appropriate time. The present invention aims to solve this problem by more accurately monitoring a user's mental health status and providing support at the appropriate time.

[1304] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1305] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for detecting risk factors from the preprocessed data, means for providing an alert to the user based on the detected risk factors, means for receiving feedback from the user, means for learning and improving the system based on the feedback, means for evaluating the user's emotions using an emotion analysis model, and means for analyzing biosensor data to detect abnormal patterns. This makes it possible to comprehensively analyze a variety of user data, detect risks early, and provide support at an appropriate time.

[1306] "Means for collecting data from users" refers to devices or software systems for obtaining users' text messages, social media posts, sensor data, etc.

[1307] "Means for pre-processing collected data" refers to processes or algorithms used to organize, filter, tokenize, etc. the data obtained.

[1308] "Risk factor detection tools" are algorithms or models that analyze pre-processed data and assess suicide risk based on specific patterns or keywords.

[1309] An "alert providing means" is a device or system for sending a warning or caution to a user based on a detected risk factor.

[1310] The "means for receiving feedback from users" refers to an interface or data collection system for collecting ratings and comments provided by users.

[1311] "Means of learning and improving the system" refers to the process of updating AI models and algorithms based on collected feedback to improve their accuracy.

[1312] The "means for evaluating emotions using an emotion analysis model" is a system that extracts and evaluates emotions from text data using a natural language processing model or other analysis tools.

[1313] "Means for analyzing biometric sensor data to detect abnormal patterns" refers to algorithms and analytical tools that analyze biometric data such as heart rate and sleep patterns to detect patterns that are out of the ordinary.

[1314] The present invention relates to a system for monitoring a user's mental health status, detecting risks early, and providing support at an appropriate time. The system is composed of the following main components:

[1315] 1. Data collection methods:

[1316] It collects users' text messages, social media posts, and biometric sensor data (e.g., heart rate, sleep patterns) using devices such as smartphones, smart glasses, and smart watches.

[1317] 2. Data preprocessing methods:

[1318] The server stores the collected data in temporary storage. It then filters out incomplete data and noise and tokenizes the text data. It uses a natural language processing library (e.g., TextBlob) to perform morphological analysis of the data and extract useful information. It also filters and cleans the sensor data in the same way.

[1319] 3. Risk factor detection methods:

[1320] The preprocessed data is input into a natural language processing model (e.g., an NLP model). This model analyzes emotions from the text and calculates an emotion score. It also analyzes biometric sensor data to detect abnormal patterns (e.g., high heart rate or abnormal sleep patterns). The server comprehensively evaluates this information and identifies risk factors for suicide.

[1321] 4. Alert Delivery Method:

[1322] If a risk factor is detected, the server will send an alert to the user's device, which will include a warning message, a chat link with a counselor, and emergency contact information, allowing the user to receive immediate and appropriate assistance.

[1323] 5. Feedback collection methods:

[1324] After providing the assistance, the user provides feedback on whether the assistance provided was appropriate, and the terminal transmits this feedback information to the server.

[1325] 6. System Learning and Improvement Measures:

[1326] The server retrains the AI ​​model based on feedback collected from users to improve the accuracy of the system.

[1327] Example

[1328] For example, if User A sends a message saying, "I've been feeling down lately," this message is collected from a smartphone app. The server tokenizes this message and performs sentiment analysis using a natural language processing model. If the message is determined to indicate a high risk factor, the server immediately sends an alert to User A's smartphone saying, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[1329] Prompt Sentence Examples

[1330] When a user sends a message saying, "I've been feeling down lately," the emotion score detected from this message is 0.7, indicating a very negative state. The user's heart rate data also shows abnormalities, indicating that immediate assistance is required.

[1331] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1332] Step 1:

[1333] A user installs the application and creates a new account. The user's smartphone, smart glasses, or smartwatch becomes a device for collecting text messages, social media posts, and biometric sensor data. This collects user input data (e.g., a text message saying "I've been feeling down lately") and sensor data (e.g., heart rate, sleep patterns).

[1334] Input: User text messages, social media posts, biometric sensor data

[1335] Output: Raw data sent to the server

[1336] Step 2:

[1337] The server stores the collected data in temporary storage. Next, it performs preprocessing, filtering out incomplete data and noise and tokenizing the text data. Specifically, it uses a natural language processing library (e.g., TextBlob) to perform morphological analysis and extract useful information. It also cleans the sensor data in a similar way, removing incomplete values.

[1338] Input: Collected raw data

[1339] Output: Filtered and tokenized preprocessed data

[1340] Step 3:

[1341] The server inputs the preprocessed text data into a natural language processing model (NLP model). The NLP model performs sentiment analysis and calculates a sentiment score for the text (e.g., -0.7). At the same time, it also detects abnormal patterns in the sensor data (e.g., a heart rate above 100). This allows for a comprehensive calculation of suicide risk.

[1342] Input: Preprocessed text data and sensor data

[1343] Output: Sentiment score and abnormal pattern detection results

[1344] Step 4:

[1345] The server evaluates risk factors based on the emotion score and abnormal patterns in the sensor data. If the risk factor is determined to be high (e.g., emotion score -0.7 or heart rate above 100), an alert is generated. The generated alert includes a warning message, a link to counseling services, and emergency contact information.

[1346] Input: Sentiment scores and anomaly pattern detection results

[1347] Output: Generated alerts

[1348] Step 5:

[1349] The server then sends the generated alert to the user's device. The user immediately receives the alert on their smartphone or smart glasses and checks the notification. The notification includes a warning such as, "Your recent messages have caused you concern. Counseling services are available," along with emergency contact information.

[1350] Input: Generated alert

[1351] Output: Notification displayed on the user's device

[1352] Step 6:

[1353] The user provides feedback on whether the provided assistance was appropriate. The terminal sends this feedback information to the server, which uses it to improve the system.

[1354] Input: User feedback

[1355] Output: Feedback information sent to the server

[1356] Step 7:

[1357] The server uses the collected feedback information to retrain the AI ​​model and improve the accuracy of the system, which will improve the accuracy and speed of future risk detection.

[1358] Input: Feedback information

[1359] Output: Retrained AI model

[1360] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1361] MODE FOR CARRYING OUT THE INVENTION

[1362] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects a variety of data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the system's accuracy. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it enables more accurate detection of risk factors. Detailed embodiments of the system are described below.

[1363] How we collect your data

[1364] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[1365] Data Preprocessing

[1366] The server stores the received data in temporary storage. Then, data preprocessing begins. The server performs the following preprocessing:

[1367] Filter out incomplete data and noise.

[1368] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library to extract useful information.

[1369] Emotion recognition by emotion engine

[1370] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[1371] Categorize emotions as negative, positive, or neutral.

[1372] A score is generated for each emotion category.

[1373] Risk factor detection

[1374] The server comprehensively analyzes the emotion scores output from the emotion engine and the sensor data. Specifically, it performs the following steps:

[1375] Keyword analysis: Detect specific risky keywords.

[1376] Sentiment analysis: Evaluating the sentiment of text data and generating a positive, negative, or neutral score.

[1377] Behavioral pattern detection: Analyze sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[1378] Calculating the overall risk score

[1379] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score, and generates an emergency alert if the overall risk score exceeds a certain threshold.

[1380] Providing alerts

[1381] The device notifies the user of the alert received from the server. The notification appears as a pop-up and contains the following:

[1382] A message expressing concern

[1383] Chat link with a counselor

[1384] Emergency contact information

[1385] Gathering feedback and training the system

[1386] The user checks the notification and enters feedback into the device on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and enables it to provide more accurate alerts and assistance in the future.

[1387] Specific examples

[1388] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the message is determined to indicate a high risk factor, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[1389] As described above, this system is capable of detecting the user's mental state early and providing support at the appropriate time.

[1390] The processing flow will be explained below.

[1391] Step 1:

[1392] Users download the app and create a new account by entering basic information (such as name, age, and gender). They then grant permission for access to text messages, social media posts, and sensor data.

[1393] Step 2:

[1394] After obtaining user consent, the device periodically collects text messages and social media posts, as well as sensor data (e.g., heart rate, steps taken, and sleep duration). This data is temporarily stored on the device.

[1395] Step 3:

[1396] The device sends the collected data to a server, where it is encrypted to protect the user's privacy.

[1397] Step 4:

[1398] The server stores the received data in temporary storage. Then, data preprocessing begins. Specifically, it performs the following processes:

[1399] Filter out incomplete data and noise.

[1400] The text data is tokenized and morphological analysis is performed using a natural language processing (NLP) library.

[1401] Step 5:

[1402] The server inputs the preprocessed text data into the emotion engine, which performs the following analysis:

[1403] Classify the sentiment of each text message or post as positive, negative, or neutral.

[1404] A score is generated for each emotion category.

[1405] Step 6:

[1406] The server uses natural language processing models to analyze the risk factors in the text data. Specifically, it performs the following tasks:

[1407] Keyword analysis: Detect specific risky keywords.

[1408] Sentiment Analysis: Evaluating the sentiment of text data and integrating the scores generated by the sentiment engine.

[1409] Step 7:

[1410] The server integrates the emotion engine score and the results of risk factor analysis of the text data to calculate an overall risk score, and generates an emergency alert if the score exceeds a certain threshold.

[1411] Step 8:

[1412] The server analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Examples include:

[1413] Abnormal sleep patterns (e.g., excessive sleep deprivation or excessive sleep)

[1414] Decreased activity level

[1415] Step 9:

[1416] The server integrates the emotion score, the text data risk score, and the sensor data results to make a final risk assessment. If the overall risk score exceeds a threshold, an alert is generated.

[1417] Step 10:

[1418] The device will notify the user of the alert information received from the server. The notification will be displayed as a popup and will include the following:

[1419] A message expressing concern

[1420] Chat link with a counselor

[1421] Emergency contact information

[1422] Step 11:

[1423] Users can view the notification and choose the appropriate action, such as starting a direct chat with a counselor or calling an emergency contact.

[1424] Step 12:

[1425] The user inputs feedback into the terminal, including an evaluation of the timing and content of the alert, on whether the alert and the provided assistance were appropriate.

[1426] Step 13:

[1427] The device sends feedback to the server, which receives it and uses it to retrain the AI ​​model, improving the system's accuracy and enabling it to provide better alerts and assistance in the future.

[1428] This is the specific processing flow of this system, which combines an emotion engine, making it possible to grasp the user's emotional state more accurately and provide support at the appropriate time.

[1429] Example 2

[1430] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1431] In recent years, mental health issues such as psychological stress and depression have become a major social issue. In particular, it is important to detect suicide risk factors early and provide support at the appropriate time. However, conventional systems can sometimes have difficulty accurately understanding a user's mental state and intervening quickly. For this reason, there is a demand for a system that can analyze a variety of data from users and detect risk factors with high accuracy.

[1432] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1433] In this invention, the server includes means for collecting data from a user, means for preprocessing the collected data, means for inputting the preprocessed data into an emotion analysis engine to generate an emotion score, and means for analyzing the emotion score and sensor data to detect risk factors, thereby enabling a highly accurate analysis of the user's mental state and early detection of risk factors.

[1434] "User" refers to any individual or entity that uses the System.

[1435] "Data" refers to information collected from users, such as text messages, social media posts, and sensor data.

[1436] "Means for collection" refers to hardware and software for collecting data with the user's permission.

[1437] "Preprocessing means" refers to the process by which collected data is filtered, tokenized, and prepared for analysis.

[1438] "Sentiment analysis engine" refers to an algorithm or model for generating sentiment scores based on pre-processed text data.

[1439] "Emotion score" refers to a numerical representation of the intensity and type of emotion extracted from text data.

[1440] "Sensor data" refers to data collected from devices that measure a user's activity patterns and biometric information.

[1441] "Risk factors" refer to certain behaviors or psychological states that indicate a user is at increased risk of suicide.

[1442] An "alert" refers to a message that notifies a user or interested party that a risk factor has been detected.

[1443] "Feedback" refers to the user providing the system with their evaluation and opinion of the assistance provided.

[1444] "Learn and improve" refers to the process of retraining the system's algorithms and models based on feedback to improve performance.

[1445] The present invention relates to a system that detects suicide risk factors early and provides support at the appropriate time. This system collects various data from users and provides alerts based on the results of analyzing that data. It also collects feedback from users and learns and improves the accuracy of the system. Detailed embodiments of the system are described below.

[1446] How we collect your data

[1447] A user downloads the app and creates a new account. They enter basic information such as their name, age, and gender, and grant permission for access to text messages, social media posts, and sensor data. The device then periodically collects the user's text messages, social media posts, and sensor data (e.g., sleep patterns and activity levels).

[1448] Data Preprocessing

[1449] The server stores the received data in temporary storage, after which it performs preprocessing on the data.

[1450] Specifically, it filters out incomplete data and noise, tokenizes the text data, and performs morphological analysis using natural language processing (NLP) libraries (e.g., NLTK and spaCy) to extract useful information.

[1451] Emotion recognition by emotion engine

[1452] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google Cloud Natural Language API), which classifies sentiment into negative, positive, and neutral and generates a score for each sentiment category.

[1453] Risk factor detection

[1454] The server comprehensively analyzes the emotion score output from the emotion engine and the sensor data. Specifically, it detects specific risk keywords, evaluates the emotion of the text data, and generates a score: positive, negative, or neutral. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle.

[1455] Calculating the overall risk score

[1456] The server combines the emotion score and the scores generated from the sensor data to calculate an overall risk score. If this overall risk score exceeds a certain threshold, the server generates an emergency alert.

[1457] Providing alerts

[1458] The device will notify the user of the alert received from the server. The notification will be displayed as a popup and will include the following:

[1459] A message expressing concern

[1460] Chat link with a counselor

[1461] Emergency contact information

[1462] Gathering feedback and training the system

[1463] The user checks the notification and enters feedback on whether the provided assistance was appropriate. The device then sends this feedback information to the server. The server receives it and uses the feedback data to retrain the AI ​​model. This improves the system's performance and allows it to provide more accurate alerts and assistance in the future.

[1464] Specific examples

[1465] For example, suppose User A sends a message within the app saying, "I've been feeling down lately." The device collects this message and sends it to the server. The server tokenizes this information and performs sentiment analysis using a natural language processing model and an emotion engine. If the result is determined to have a high negative score, the server immediately sends an alert to the device, informing User A, "Your recent messages have made me worried. Counseling services are available." At this time, User A can receive appropriate support.

[1466] This allows the system to detect the user's mental state early and provide support at the appropriate time.

[1467] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1468] Step 1:

[1469] Users download the app and create a new account, entering basic information such as name, age, and gender, and granting access to text messages, social media posts, and sensor data, which is then sent by the device to a server.

[1470] Input: Name, age, gender, text messages, social media posts, sensor data

[1471] Output: Send basic information and permission information to the server

[1472] Step 2:

[1473] The server receives the user's basic information and permission information and stores it in temporary storage. The device periodically collects the user's text messages, social media posts, and sensor data and sends them to the server. The server stores the received data in temporary storage.

[1474] Input: User basic information, text messages, social media posts, sensor data

[1475] Output: Saving data to temporary storage

[1476] Step 3:

[1477] The server begins preprocessing the collected data by filtering out incomplete data and noise, tokenizing the text data, and extracting useful information from the data by performing morphological analysis using a natural language processing (NLP) library (e.g., NLTK or spaCy).

[1478] Input: Raw data stored in temporary storage

[1479] Output: Preprocessed data (filtered, tokenized, morphologically analyzed)

[1480] Step 4:

[1481] The server inputs the preprocessed text data into a sentiment analysis engine (e.g., Google Cloud Natural Language API), which generates a sentiment score for the text and classifies it into negative, positive, or neutral categories.

[1482] Input: Preprocessed text data

[1483] Output: Sentiment scores (negative, positive, neutral scores)

[1484] Step 5:

[1485] The server comprehensively analyzes the emotion score and sensor data. Specifically, it detects specific risk keywords and evaluates the emotion score. It also analyzes the sensor data to detect abnormal behavioral patterns and changes in lifestyle. Based on the results of this analysis, it generates individual scores.

[1486] Input: emotion scores, sensor data

[1487] Output: Risk factor score

[1488] Step 6:

[1489] The server combines the generated individual scores to calculate an overall risk score, and if this overall risk score exceeds a certain threshold, the server generates an emergency alert and sends it to the device.

[1490] Input: Risk factor score

[1491] Output: Overall risk score, emergency alert

[1492] Step 7:

[1493] The device notifies the user of emergency alerts received from the server. The notification appears as a pop-up and includes a message of concern, a chat link to a counselor, and emergency contact information.

[1494] Input: Emergency Alert

[1495] Output: A popup notification to the user

[1496] Step 8:

[1497] The user checks the notification and inputs feedback on whether the provided assistance was appropriate or not into the terminal, which then transmits this feedback information to the server.

[1498] Input: Feedback

[1499] Output: Sending feedback to the server

[1500] Step 9:

[1501] The server uses the received feedback data to retrain the AI ​​model, improving the system's performance and enabling it to provide more accurate alerts and assistance in the future.

[1502] Input: Feedback data

[1503] Output: Retrained AI model

[1504] (Application example 2)

[1505] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1506] In modern society, it is increasingly important to maintain users' mental health, detect risk factors early, and provide appropriate support. However, conventional systems have had difficulty properly collecting and analyzing diverse data and detecting risk factors in real time. Furthermore, a lack of means to provide support at the appropriate time has made it difficult to intervene effectively to ensure users' mental health. Therefore, the present invention aims to solve these problems by detecting users' mental states early and providing appropriate support.

[1507] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1508] In this invention, the server includes means for collecting data from the user, means for preprocessing the collected data, and means for detecting risk factors from the preprocessed data. This allows for the use of emotion scores and behavioral patterns in detecting risk factors. Furthermore, by using a means for providing notifications via a smartphone or smart glasses, it is possible to provide prompt and accurate alerts to the user. Furthermore, the system can receive feedback from the user and continue to learn and improve based on that feedback, thereby enabling continuous monitoring of the user's mental health and providing appropriate support.

[1509] "User" means an individual or organization that uses this system.

[1510] A "means for collecting data" is a device or function for transmitting user text messages, social media posts, sensor data, etc. to a server.

[1511] A "preprocessing means" is a device or function for organizing collected data, filtering out incomplete data and noise, and converting it into an analyzable format.

[1512] The "means for detecting risk factors" is a device or function for evaluating the user's mental health status and behavioral patterns from the preprocessed data and recognizing abnormal conditions and risks.

[1513] A "means for providing an alert" is a device or function for providing a warning or counseling suggestion to a user when a risk factor is detected.

[1514] A "means for receiving feedback" is a device or function for informing the server how the user responded to the alert or whether the assistance provided was adequate.

[1515] A "means for learning and improving the system" is a device or function that updates the system's algorithms based on user feedback to improve accuracy and effectiveness.

[1516] An "emotion score" is a numerical assessment of a user's emotional state based on text data and other data.

[1517] "Behavioral patterns" refer to a user's daily behavior and lifestyle habits analyzed from sensor data.

[1518] A "smartphone" is a portable communication device that has functions such as voice calls, text message sending, and internet connection.

[1519] "Smart glasses" are eyeglass-type devices incorporating display devices and sensors, and are devices that provide visual information to users.

[1520] A system for implementing this invention mainly uses the following hardware and software. The hardware uses a smartphone or smart glasses. The software includes a natural language processing (NLP) library, an emotion recognition engine, and cloud data storage. For example, "spaCy" or "NLTK" can be used as the natural language processing library. Possible emotion recognition engines include "Amazon Comprehend" or "Google Cloud Natural Language API." Furthermore, "AWS S3" or "Google Cloud Storage" is used as the cloud data storage. "React Native" or "Flutter" is used for mobile app development.

[1521] The server collects user data such as text messages, social media posts, and sensor data. During data preprocessing, the collected data is organized and incomplete data and noise are filtered. The preprocessed data is input into a natural language processing model and an emotion recognition engine. The server uses emotion scores and behavioral patterns to detect risk factors and provide alerts to users.

[1522] As a concrete example, a user may send a message via a smartphone app saying, "I'm feeling depressed." The server tokenizes the message and analyzes it using a natural language processing model and an emotion recognition engine. If the text is determined to indicate a high risk factor, the server will provide an alert to the user via their smartphone or smart glasses. The alert message may say, "Your recent messages have caused us concern. Counseling services are available."

[1523] The prompt statement in the example program is:

[1524] Input prompt: User types "I'm feeling down"

[1525] System Output: "Your recent messages have me concerned. Counseling services are available."

[1526] This allows users to receive appropriate assistance quickly. The system also learns and improves based on user feedback, which will enable more accurate risk factor detection and assistance provision in the future.

[1527] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1528] Step 1:

[1529] A user uses a smartphone or smart glasses to input a text message, for example, the user inputs the message "I'm feeling down," which is sent to the system.

[1530] Step 2:

[1531] The terminal sends the entered text message to the server, where it is temporarily stored.

[1532] Step 3:

[1533] The server preprocesses the received text message. The input is the text message, and the output is tokenized text. Specifically, it uses a natural language processing library (e.g., spaCy) to tokenize the message and perform morphological analysis.

[1534] Step 4:

[1535] The server inputs the preprocessed text data into an emotion recognition engine. The input is tokenized text, and the output is an emotion score. This process uses services such as Amazon Comprehend or Google Cloud Natural Language API to generate an emotion score (e.g., negative, positive, neutral).

[1536] Step 5:

[1537] The server combines the generated emotion scores with the sensor data to calculate an overall risk score. Specifically, it numerically evaluates and combines the emotion scores and behavioral pattern data (e.g., sleep patterns, activity levels). This results in an overall risk score.

[1538] Step 6:

[1539] If the overall risk score exceeds a certain threshold, the server generates an alert and notifies the device. Specifically, the notification content is created as a message such as "Your recent messages have caused you concern. Counseling services are available." and sent to the smartphone or smart glasses.

[1540] Step 7:

[1541] The device displays the received alert notification to the user, who can then check the alert and access information such as counseling services or emergency contacts as needed.

[1542] Step 8:

[1543] The user inputs feedback about the alert into the terminal, such as "The notified support was appropriate" or "I used the service."

[1544] Step 9:

[1545] The terminal transmits the feedback from the user to the server, which stores the feedback.

[1546] Step 10:

[1547] The server uses the feedback it receives to learn and improve the system. Specifically, it inputs the feedback data into a generative AI model, retraining the algorithm and improving the overall system performance, which will enable more accurate risk factor detection and assistance provision in the future.

[1548] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1549] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1550] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1551] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1552] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1553] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1554] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1555] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1556] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1557] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1558] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1559] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1560] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1561] 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.

[1562] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1563] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1564] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1565] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1566] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1567] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1568] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1569] The following is further disclosed regarding the above embodiment.

[1570] (Claim 1)

[1571] a means for collecting data from a user;

[1572] means for pre-processing the collected data;

[1573] means for detecting risk factors from the preprocessed data;

[1574] means for providing an alert to a user based on the detected risk factors;

[1575] a means for receiving feedback from users;

[1576] A system that includes a means to learn and improve the system based on feedback.

[1577] (Claim 2)

[1578] 10. The system of claim 1, wherein the system collects user text messages, social media posts, and sensor data.

[1579] (Claim 3)

[1580] 10. The system of claim 1, wherein the preprocessed data is input into a natural language processing model.

[1581] "Example 1"

[1582] (Claim 1)

[1583] a means for collecting information from a user;

[1584] a means for temporarily storing the collected information;

[1585] a means for cleansing and tokenizing the temporarily stored information;

[1586] a means for analyzing the tokenized information using natural language processing;

[1587] A means for detecting risk factors from the analysis results and calculating a risk score;

[1588] means for providing a warning to a user when the calculated risk score exceeds a threshold;

[1589] a means for receiving feedback from users;

[1590] A system that includes a means to learn and improve the system based on feedback.

[1591] (Claim 2)

[1592] 10. The system of claim 1, wherein the system collects user text messages, social media posts, and activity data.

[1593] (Claim 3)

[1594] 10. The system of claim 1, wherein the preprocessed information is input into a generative AI model.

[1595] "Application Example 1"

[1596] (Claim 1)

[1597] a means for collecting data from a user;

[1598] means for pre-processing the collected data;

[1599] means for detecting risk factors from the preprocessed data;

[1600] means for providing an alert to a user based on the detected risk factors;

[1601] a means for receiving feedback from users;

[1602] a means to learn and improve the system based on feedback;

[1603] means for assessing a user's emotions using an emotion analysis model;

[1604] A system including means for analyzing biosensor data to detect abnormal patterns.

[1605] (Claim 2)

[1606] The system of claim 1, characterized in that it collects user text messages, social media posts, and sensor data and performs sentiment analysis.

[1607] (Claim 3)

[1608] The system according to claim 1, characterized in that the preprocessed data is input into a natural language processing model, a prompt sentence is generated, and an alert notification is sent.

[1609] "Example 2: Combining Emotion Engines"

[1610] (Claim 1)

[1611] a means for collecting data from a user;

[1612] means for pre-processing the collected data;

[1613] means for inputting the preprocessed data into a sentiment analysis engine to generate a sentiment score;

[1614] a means for analyzing the emotion score and the sensor data to detect risk factors;

[1615] means for providing an alert to a user based on the detected risk factors;

[1616] a means for receiving feedback from users;

[1617] A system that includes a means to learn and improve the system based on feedback.

[1618] (Claim 2)

[1619] 10. The system of claim 1, wherein the system collects user text data, social media posts, and sensor data.

[1620] (Claim 3)

[1621] 2. The system according to claim 1, wherein the preprocessed data is input into a natural language processing engine to perform morphological analysis.

[1622] "Application example 2 when combining emotion engines"

[1623] (Claim 1)

[1624] a means for collecting data from a user;

[1625] means for pre-processing the collected data;

[1626] means for detecting risk factors from the preprocessed data;

[1627] means for providing an alert to a user based on the detected risk factors;

[1628] a means for receiving feedback from users;

[1629] a means to learn and improve the system based on feedback;

[1630] a means for using emotion scores and behavioral patterns in detecting risk factors;

[1631] A system including means for providing notifications using a smartphone or smart glasses.

[1632] (Claim 2)

[1633] 10. The system of claim 1, wherein the system collects user text messages, social media posts, and sensor data.

[1634] (Claim 3)

[1635] 10. The system of claim 1, wherein the preprocessed data is input to a natural language processing model and an emotion recognition engine. [Explanation of symbols]

[1636] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for collecting data from a user; means for pre-processing the collected data; means for detecting risk factors from the preprocessed data; means for providing an alert to a user based on the detected risk factors; a means for receiving feedback from users; A system that includes a means to learn and improve the system based on feedback.

2. The system of claim 1 , wherein the system collects user text messages, social media posts, and sensor data.

3. The system of claim 1 , wherein the preprocessed data is input to a natural language processing model.

Images