System

Abstract

A system is provided.SOLUTION: A system comprising: means for monitoring users' online behavior; means for analyzing users' behavior and evaluating risk levels; means for displaying warnings for risky behavior; means for providing users with predicted damage and contact information; and means for recording users' reactions and updating AI models.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In recent years, crimes and harms have become more diverse in the use of the Internet, with an increasing number of victims, particularly among elderly people and minors with low internet literacy. In this situation, there is a need to provide an environment where users can safely and securely engage in online activities. However, current security systems have limited ability to detect specific behaviors and risks, and are unable to provide users with timely warnings. Furthermore, there are insufficient systems that can predict risks and suggest appropriate countermeasures before harm occurs. Therefore, there is a need for effective means for users to proactively avoid the risks of online crime and fraud. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means. The system includes a means for monitoring users' online behavior, a means for analyzing users' behavioral data and assessing risk levels, a means for displaying warnings for risky behavior, a means for providing users with estimated damage amounts and information on where to seek help, and a means for recording users' responses and updating an AI model. This system uses a generative model to learn from past crime data and perform risk analysis, allowing it to analyze users' behavioral data in real time and display immediate warnings. This allows users to prevent harm before it occurs and engage in online activities with peace of mind.

[0006] "User" means any person or entity that uses the Online Services.

[0007] "Online behavior" refers to activities such as searching, posting, and communicating on the Internet.

[0008] "Monitoring measures" refers to technologies and devices that capture and record users' online behavior in real time.

[0009] "Behavioral Data" means information about the actions you take on the Internet.

[0010] "Means for analyzing and assessing risk levels" refers to algorithms and technologies for determining risk based on acquired behavioral data.

[0011] "Means for displaying a warning" means a technology or device that displays a warning message to a user about risky behavior.

[0012] "Expected damage amount" refers to the expected financial loss if you become a victim of fraud or crime.

[0013] "Consultation information" means contact information for institutions or organizations that users can consult if they are victims of fraud or crime.

[0014] A "generative model" refers to a learning model that uses AI to perform risk analysis from past crime data.

[0015] "Risk analysis" refers to the process of assessing the likelihood of fraud or crime based on user behavior data.

[0016] "Displaying an immediate warning" means instantly issuing a warning to the user when a risk is detected. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] The following describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive service for reducing the risk of users becoming involved in crimes or becoming victims online, and specifically operates in the following manner.

[0039] System Configuration

[0040] The system mainly consists of the following components:

[0041] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral data.

[0042] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[0043] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0044] Program processing and operation explanation

[0045] Users begin online activities

[0046] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[0047] Behavioral Data Capture

[0048] The server captures user behavior data in real time. To protect user privacy, it does not collect data that can identify personal information. User search queries, postings, received emails, etc. are monitored.

[0049] Risk Analysis

[0050] The captured behavioral data is input into a generative model stored on the server. The generative model learns from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if the search query "cheapest brand name products" is determined to be likely to be related to fraud, it will be classified as high risk.

[0051] Displaying warnings

[0052] If the risk is deemed high, the server will generate an appropriate warning message for the user, which will be displayed on the user's device. For example, the message may read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0053] Providing estimated damage amounts and information on where to seek advice

[0054] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0055] User Reactions and Feedback

[0056] The server records whether the user ignores the warning or whether they act on it, and the recorded data is used as training data for the generative model and fed back to improve the accuracy of the entire system.

[0057] Specific examples

[0058] Example 1: If a user posts on social media, "I want to buy expensive brand-name goods at a discount," the server will scan the content and detect risks. If applicable, it will display a warning and provide an estimated loss amount and information on where to seek advice.

[0059] Example 2: When a user receives an email containing a suspicious link, the server analyzes the email and displays a warning if it determines that the risk is high. It also notifies the user of the damage that could result from clicking the link and urges them to take action immediately.

[0060] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0061] The processing flow will be explained below.

[0062] Step 1:

[0063] The user begins their online journey. They use their device to access the Internet and search, write, and communicate. For example, they enter a search query such as "cheapest brand name products."

[0064] Step 2:

[0065] The device captures the user's online behavior, including search queries, social media posts, and received emails, and sends this data to a server.

[0066] Step 3:

[0067] The server receives the behavioral data sent from the device in real time, where it undergoes pre-processing and privacy-conscious data filtering.

[0068] Step 4:

[0069] The server inputs the received data into an internal generative model, which has learned from past crime data and fraud patterns to assess the risk level. For example, it analyzes search results for "cheapest brand name products" and determines that there is a high probability of fraud.

[0070] Step 5:

[0071] The server receives the evaluation results from the generative model and determines the risk level. It is classified as high risk, medium risk, or low risk. If there is a high possibility of fraud, it is classified as high risk.

[0072] Step 6:

[0073] If the server determines that the risk is high, it will generate an appropriate warning message for the user, which will include specific risks and countermeasures.

[0074] Step 7:

[0075] The server generates a warning message and sends it to the user's device, such as "This search result may contain fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0076] Step 8:

[0077] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[0078] Step 9:

[0079] The server generates a warning message along with estimated damages and information on where to seek advice, and provides it to the user. For example, it might say, "If you fall victim to a scam, you are likely to lose 200,000 yen. Please contact the following organization for advice."

[0080] Step 10:

[0081] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[0082] Step 11:

[0083] The server records how the user responded to the warning, for example, by storing a log of whether the warning was ignored, acted upon, or whether a helpline was contacted.

[0084] Step 12:

[0085] The server uses the recorded data as training data for the generative model, and the feedback improves the accuracy of the generative model, adjusting it to make future risk analyses more accurate.

[0086] Through the above process, the Digital Lifeguard System effectively reduces the risk of online crime and fraud, providing users with an environment where they can use the Internet with peace of mind.

[0087] Example 1

[0088] 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."

[0089] As the number of Internet users increases daily, the risk of becoming involved in online crimes and fraud is also increasing. In particular, many users suffer financial losses due to clicking on fraudulent links or visiting fraudulent websites. However, existing security measures have difficulty monitoring users' online behavior in real time and immediately warning them of risks, so effective countermeasures are needed.

[0090] 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.

[0091] In this invention, the server includes means for monitoring users' online behavior, means for capturing and preprocessing user behavior data in real time, means for inputting the captured data into a generative AI model and evaluating the risk level, means for displaying a warning for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the generative AI model. This provides an environment in which users can act safely online, and has the effect of significantly reducing the risk of crime and fraud.

[0092] "Online behavior" refers to a series of activities that users perform on the Internet, such as searching, posting, and receiving emails.

[0093] "Behavioral Data" is data that contains information about your online behavior, such as search queries, posts, and received emails.

[0094] "Real-time" means that data is processed as soon as it is generated, allowing for immediate analysis and evaluation without delay.

[0095] "Capture" is the act of collecting data about a user's online behavior.

[0096] "Preprocessing" is the process of removing unnecessary information from the captured raw data and converting it into a form suitable for analysis.

[0097] A "generative AI model" is an artificial intelligence algorithm that learns from past crime data and assesses risk based on new data.

[0098] A "risk level" is a number or category that evaluates the degree of danger associated with a user's behavior.

[0099] A "warning message" is a notification that alerts users to risky behavior.

[0100] "Estimated damage amount" is the amount of financial loss that a user is expected to incur if they fall victim to fraud or crime.

[0101] "Consultation information" is contact information for institutions or organizations that users should contact if they become victims.

[0102] "Recording a response" means recording how the user responded to the warning (whether they ignored it or followed it).

[0103] "Updating the AI ​​model" is the process of relearning the artificial intelligence algorithm based on recorded reaction data to improve its accuracy.

[0104] MODE FOR CARRYING OUT THE INVENTION

[0105] This invention is a system for reducing the risk of being involved in online crime and fraud by monitoring users' online behavior, assessing risks, and providing appropriate warning messages. The system mainly consists of three main components: a server, a user's device, and a generative AI model.

[0106] System Configuration

[0107] 1. Server:

[0108] The server captures and preprocesses users' online behavioral data in real time. It is preferable to use cloud services such as AWS (Amazon Web Services) or Google Cloud Platform. The server also contains a generative AI model built using machine learning libraries such as TensorFlow and PyTorch. This model evaluates risk levels based on the captured data.

[0109] 2. Terminal:

[0110] A terminal is a device that a user uses to access the internet, such as a PC, smartphone, or tablet. The terminal is responsible for displaying warning messages and consultation information sent from the server.

[0111] 3. Generative AI Model:

[0112] Generative AI models are trained using machine learning libraries such as TensorFlow and PyTorch based on historical crime data and fraud patterns, and then analyze captured user behavior data to assess risk levels.

[0113] Specific examples

[0114] Example 1: Risk assessment of search queries

[0115] When a user searches for "cheapest branded goods" on their device, the server captures the search query in real time. The captured data is input into a generative AI model, which assesses the risk based on past fraud patterns. If the risk is determined to be high, the server generates a warning message that reads, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers." and displays it on the device.

[0116] Example 2: Analysis of suspicious emails

[0117] When a user receives an email containing a suspicious link, the server captures the email in real time and performs text analysis. If the generative AI model determines that the email poses a high risk, the server generates a warning message that reads, "This email may be fraudulent. Do not click on the link." and displays it on the device.

[0118] Prompt Sentence Examples

[0119] "Analyze current user behavior based on past fraud data, conduct risk assessments, and generate warning messages if deemed high risk. Also consider countermeasures if users ignore the messages."

[0120] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0121] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0122] Step 1:

[0123] A user accesses the Internet. Using a device (PC, smartphone, tablet, etc.), the user opens a web browser and searches for, for example, "cheapest brand-name products."

[0124] Input: search query "cheapest brand name products".

[0125] Output: The user's search query is retrieved.

[0126] Step 2:

[0127] The server captures user behavior data in real time, including the search queries users type, the content they post, and the emails they receive.

[0128] Input: Your online behavior data (search queries, posts, received emails, etc.).

[0129] Output: Captured behavioral data.

[0130] Step 3:

[0131] The server preprocesses the captured behavioral data. This involves filtering unnecessary information from the text data and converting it into a format suitable for analysis. For example, natural language processing (NLP) techniques are used to extract important keywords.

[0132] Input: Raw captured data.

[0133] Output: Preprocessed data (data from which important keywords have been extracted).

[0134] Step 4:

[0135] The server then inputs the preprocessed data into a generative AI model, which uses historical crime data and fraud patterns to assess risk levels. For example, it might determine that a query like "cheapest brand name products" is likely to be fraud-related.

[0136] Input: Preprocessed data.

[0137] Output: Risk level assessment result (high risk, low risk, etc.).

[0138] Step 5:

[0139] If the server determines that the risk is high, it generates a warning message, which is then displayed on the user's device. For example, the message might read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0140] Input: Risk level assessment result (high risk).

[0141] Output: A warning message.

[0142] Step 6:

[0143] The server calculates the estimated loss if you fall victim to fraud or crime and provides it to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0144] Input: Risk level assessment result (high risk).

[0145] Output: Estimated damage amount and consultation information.

[0146] Step 7:

[0147] The server records whether the user ignored or acted on the warning. This record includes the user's actions after the warning (whether they clicked on a link, acted on the warning, etc.). The recorded data is used as training data for the generative AI model.

[0148] Input: User response data.

[0149] Output: An updated generative AI model.

[0150] In this way, the system monitors users' online behavior in real time and issues alerts when necessary, reducing the risk of crime and fraud.

[0151] (Application example 1)

[0152] 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."

[0153] In modern society, the number of internet-based frauds and crimes is on the rise, creating an urgent need to provide an environment where users can safely engage online. However, current security services lack real-time monitoring of online behavior and immediate warnings, limiting their effectiveness in preventing damage. Furthermore, they lack a feedback mechanism that allows the system to evolve based on user responses, leaving challenges in improving the overall accuracy of the system. To address these challenges, a comprehensive system with real-time monitoring, immediate warnings, and a feedback mechanism is needed.

[0154] 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.

[0155] In this invention, the server includes means for monitoring users' online behavior, means for analyzing user behavior data and assessing risk levels, means for displaying warnings for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, means for recording users' responses and updating the generative model, means for monitoring user behavior in real time, and means for analyzing risks in real time and displaying immediate warnings. This provides an environment where users can engage in online activities with peace of mind, making it possible to significantly reduce the risk of crime and fraud.

[0156] "User" means any person or entity that uses the System to conduct online activities.

[0157] "Online behavior" is a general term for actions such as searching, posting, and communication that are carried out using the Internet.

[0158] A "server" is a central component of the system and is an information processing device that collects, analyzes, and stores user behavior data.

[0159] "Behavioral Data" is a record of certain actions you take online, including search queries, postings, and the content of emails you receive and send.

[0160] "Risk level" is an indicator that shows the likelihood that a user's behavior will lead to crime or fraud.

[0161] A "warning" is a message that warns users against high-risk behavior.

[0162] "Estimated damage amount" is the amount of financial damage expected if a user falls victim to fraud or crime.

[0163] "Consultation information" refers to contact information for the police, consumer centers, etc. that users can consult if they are victims of a crime or fraud.

[0164] A "generative model" is an artificial intelligence model that learns from past crime data and is used to analyze the risks associated with user behavior.

[0165] A "feedback mechanism" is a mechanism for recording user responses and using that data to improve the accuracy of the generative model.

[0166] "Real-time monitoring" refers to the monitoring of users' online behavior in real time.

[0167] "Immediate warning" means that a warning message is displayed immediately when high-risk behavior is detected.

[0168] MODE FOR CARRYING OUT THE INVENTION

[0169] This paper describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive security service that reduces the risk of users becoming involved in online crimes or becoming victims. Below, each component of the system and its operation will be specifically described.

[0170] System Configuration

[0171] The system mainly consists of the following components:

[0172] 1. Server: This is the main component that monitors users' online behavior and analyzes the behavioral data.

[0173] 2. Terminal: A device (personal computer, smartphone, tablet, etc.) that a user uses to access the Internet.

[0174] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0175] Program processing and operation explanation

[0176] Monitoring your online behavior

[0177] Users access the Internet using their devices to search, post, and communicate. For example, suppose a user searches for "cheapest brand-name products." All of this behavior is monitored in real time.

[0178] Behavioral data capture and analysis

[0179] The server captures user behavior data in real time and inputs this data into a generative model. The generative model has learned from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if a search query such as "cheapest brand name products" is determined to be highly likely to be related to fraud, it will be classified as high risk.

[0180] Risk analysis and warning display

[0181] If the risk is determined to be high in real time, the server generates an appropriate warning message for the user and displays it on the device, such as "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from an authorized retailer."

[0182] Providing estimated damage amounts and information on where to seek advice

[0183] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0184] User Reactions and Feedback

[0185] The server records whether the user ignores the warning or whether they act on it. The recorded data is used as training data for the generative model and as feedback to improve the accuracy of the entire system.

[0186] Hardware and software used

[0187] Hardware: Personal computers, smartphones, tablets

[0188] software:

[0189] AI Library: Used to build generative AI models and perform risk analysis

[0190] Monitoring Tools: Monitor your users' online behavior with UserActionMonitor

[0191] Notification Management System: Display alerts using NotificationManager

[0192] Examples and prompts

[0193] Specific examples

[0194] If a user posts on social media that they want to buy expensive branded goods at a discount, the server will scan the post and detect any risks. If applicable, it will display a warning, provide an estimated loss amount, and provide information on where to seek advice.

[0195] Prompt Sentence Examples

[0196] "When a user searches for 'cheapest branded goods,' please generate a prompt to check for fraud risk."

[0197] As described above, the digital lifeguard system according to the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0198] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0199] Step 1:

[0200] A user accesses the Internet using a device. This initiates online activities such as search queries, posting, and receiving / sending emails. The user's online behavior data (search queries, social media postings, email content) is obtained as input. This becomes the initial input for the system.

[0201] Step 2:

[0202] The server monitors the user's online behavior in real time. Here, we use the UserActionMonitor tool to capture the user's behavior data. As input, the user's online behavior data is monitored and sent to the server as behavior data. As output, this data is saved as behavior data capture.

[0203] Step 3:

[0204] The server inputs the captured behavioral data into a generative AI model. This AI model has learned from past crime data and uses an algorithm to assess risk levels. The captured behavioral data is passed to the generative AI model as input. The output is an assessment of the risk level for the behavioral data.

[0205] Step 4:

[0206] The server makes a decision based on the risk level assessed by the generative AI model. A warning message is generated for actions with a high risk level. Here, the WarningGenerator tool is used. The input is the risk level assessment result. The output is a generated warning message.

[0207] Step 5:

[0208] The server notifies the user's terminal of the generated warning message. Here, we use the NotificationManager tool. As input, there is the generated warning message, which is sent to the user's terminal. As output, the warning message is displayed on the user's terminal.

[0209] Step 6:

[0210] The server calculates and provides estimated damage amounts and information on where to seek advice for high-risk behavior. The inputs are behavioral data and risk level assessment results. The output is a calculated damage amount and information on where to seek advice, which are provided to the user.

[0211] Step 7:

[0212] It monitors and records how users react to warnings and provided information. The server collects this reaction data and uses it as training data for the generative model. As input, user reaction data is collected. As output, this data is saved as feedback for the generative model.

[0213] These steps enable the Digital Lifeguard system to monitor users' online behavior in real time and provide appropriate warnings and information.

[0214] 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.

[0215] This paper describes a digital lifeguard system incorporating an emotion engine, which is an interactive service for users to reduce the risk of becoming involved in online crimes and victimization. The system also recognizes the user's emotional state and provides warnings and advice based on that state.

[0216] System Configuration

[0217] The system mainly consists of the following components:

[0218] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral and emotional data.

[0219] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[0220] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0221] 4. Emotion Engine: A component that recognizes and analyzes user emotions in real time.

[0222] Program processing and operation explanation

[0223] Users begin online activities

[0224] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[0225] Capturing behavioral and emotional data

[0226] The device captures the user's online behavior and input from the camera and microphone. The emotion engine analyzes the user's emotions using the captured video and audio data. This data is sent to the server.

[0227] Risk and Sentiment Analysis

[0228] The server receives and analyzes the behavioral and emotional data. The behavioral data is input into a generative model to assess the risk level. The emotional data is analyzed by an emotion engine to assess the user's current emotional state (e.g., stress, relief, anxiety).

[0229] Generating Emotion-Based Alerts

[0230] The server generates an appropriate warning message based on the risk level and the evaluation of the user's emotional state. For example, if the user is feeling stressed, the server can issue a warning in a gentle tone that takes into account the user's psychological state. A specific example of a message that could be generated is, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time."

[0231] Displaying warnings and providing countermeasure information

[0232] The server sends the generated warning message to the user's device. The device displays the warning message in real time, along with the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization for advice."

[0233] User Reactions and Feedback

[0234] The server records how users respond to the alerts and how their emotional state changes. These data are used as training data for the generative model and emotion engine, and are fed back to improve the accuracy of the entire system.

[0235] Specific examples

[0236] Example 1: If a user posts on social media, "I want to buy expensive branded goods at a discount," the server will scan the content and the user's emotions to detect the risk of fraud. Also, if the user's emotional state is "excited," the server will display a warning in a gentle tone to help them regain their composure.

[0237] Example 2: If a user receives an email containing a suspicious link, the server analyzes the email and, if it detects that the user's emotional state is "anxiety," it suggests ways to mitigate the risk and provides a message to help the user stay calm.

[0238] In this way, the digital lifeguard system of the present invention, which is combined with an emotion engine, more effectively provides an environment in which users can safely engage online, and realizes risk management that takes into account their emotional state.

[0239] The processing flow will be explained below.

[0240] Step 1:

[0241] The user begins their online journey. They use their device to access the Internet and search, post, and communicate. For example, they search for "cheapest brand name products."

[0242] Step 2:

[0243] The device captures the user's online behavior, such as search queries, social media posts, and received emails, and sends this data to a server.

[0244] Step 3:

[0245] The device's camera and microphone are used to capture emotional data from the user's facial expressions and voice, and this data is also sent to the server.

[0246] Step 4:

[0247] The server receives behavioral and emotional data sent from the device in real time, and performs privacy-conscious data filtering at this stage.

[0248] Step 5:

[0249] The server inputs the received behavioral data into a generative model, which is trained on historical crime data and fraud patterns and uses this data to assess risk levels. For example, a search query for "cheapest branded goods" might be deemed to have a high risk of fraud.

[0250] Step 6:

[0251] The server analyzes the emotion data using an emotion engine, evaluating the user's emotional state based on their facial expressions and voice, and detecting, for example, "anxiety" or "excitement."

[0252] Step 7:

[0253] The server combines the risk level assessment results with the emotional state assessment results to generate an appropriate warning message. If the risk is high and the user is feeling "anxious," it creates a gentler warning that takes into consideration the user's feelings. For example, it might say, "This search result may include fraudulent sites selling counterfeit branded goods. Please check with authorized retailers when you have time."

[0254] Step 8:

[0255] The server sends the generated warning message to the user's device. At the same time, it also generates an estimated loss amount and information on where to seek advice. For example, "If you fall victim to a scam, you can expect to lose 200,000 yen. Please contact the following organizations for advice."

[0256] Step 9:

[0257] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[0258] Step 10:

[0259] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[0260] Step 11:

[0261] The server records how the user responded to the warning, for example, whether the warning was ignored or acted upon, and also records the user's emotional state in response to the warning.

[0262] Step 12:

[0263] The server uses the recorded behavioral and emotional data as training data for the generative model and emotion engine, improving the accuracy of the entire system and adjusting it to ensure more accurate risk and emotion analysis from the next time onwards.

[0264] Through the above process, by combining an emotion engine with the digital lifeguard system, user safety can be ensured more effectively.

[0265] Example 2

[0266] 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."

[0267] In today's digital society, users are at a greater risk of becoming involved in crimes and fraud online. Furthermore, warnings and advice that ignore users' emotional state make it difficult to encourage appropriate responses. As a result, users' safety may not be adequately ensured, and the damage may worsen.

[0268] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing user behavioral data and emotional data and transmitting them to the server, means for inputting the user behavioral data into a generative model to evaluate the risk level, and means for analyzing the emotional data using an emotion engine to evaluate the user's emotional state. This ensures the safety of the user and enables appropriate warnings and advice that take into account the user's emotional state.

[0269] "Means for monitoring users' online behavior" refers to the ability to track and record users' activities on the Internet, such as searches, posts, and clicks, in real time.

[0270] "Means for capturing user behavioral data and emotional data and transmitting it to the server" refers to a function that collects the user's operation history, facial expressions, tone of voice, etc., and transfers this data to the server.

[0271] The "means of inputting into a generative model to assess risk levels" refers to a function that inputs acquired behavioral data into an AI model and performs risk assessment based on past crime data.

[0272] "Means for analyzing emotional data using an emotion engine and assessing the user's emotional state" is a function that analyzes captured video and audio data and determines the user's emotional state.

[0273] "Means for generating an appropriate warning message based on the risk level and emotional state and sending it to the user's device" refers to a function that creates an appropriate warning message for the user based on the risk assessment and emotional state, and immediately sends it to the user's device.

[0274] "Means for providing users with estimated damage amounts and information on consultation centers" is a function that provides users with information on appropriate consultation centers along with estimated damage amounts when a risk such as fraud is detected.

[0275] "Means for recording user responses and updating the AI ​​model" refers to a function that records user responses to warning messages and uses that data to improve the accuracy and efficiency of the AI ​​model.

[0276] A "generative model" is an artificial intelligence model that learns from past crime data and compares it with current data to perform risk assessment.

[0277] An "emotion engine" is a system element that analyzes emotional data such as images and sounds and determines the user's emotional state.

[0278] The digital lifeguard system of this invention ensures the safety of users online and provides appropriate warnings and advice taking into account their emotional state. This system mainly consists of four components: a server, a terminal, a generative model, and an emotion engine.

[0279] The server is the central part of the system and plays a key role in analyzing users' behavioral and emotional data. The terminal is a device (such as a PC, smartphone, or tablet) that users use to access the Internet, and is responsible for capturing behavioral and emotional data and sending it to the server. The generative model is installed in the server and is an AI model used to perform risk analysis based on past crime data. The emotion engine is a component that recognizes and analyzes users' emotions in real time.

[0280] The system begins operation when a user accesses the Internet using a device and begins searching, posting, or communicating. The device captures the user's online behavior and simultaneously captures input data from a camera and microphone. For example, the device's camera captures the user's facial expressions and the microphone records the user's tone of voice. This data is analyzed by an emotion engine to evaluate the user's emotional state.

[0281] The server receives the behavioral and emotional data sent from the device and uses it to analyze each. The behavioral data is input into a generative model to evaluate the risk level. For example, if a user attempts to click on a suspicious link, it is compared with a past database to determine whether the risk is high. The emotional data is analyzed by an emotion engine to evaluate the user's current emotional state. For example, it analyzes whether the user is feeling stressed while searching.

[0282] The server generates an appropriate warning message based on the risk level and the user's emotional state. If the user is feeling stressed, the server can issue a gentle warning that takes into account the user's psychological state. For example, a message such as "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time" can be generated.

[0283] The server sends the generated warning message to the user's device, which displays it in real time. It also displays the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization."

[0284] The server records how the user responded to the warning and also records how their emotional state changed. This data is used as training data for the generative model and emotion engine, and is fed back to improve the accuracy of the entire system.

[0285] Examples of prompt statements

[0286] Examples of prompts include:

[0287] 1. "Generate warning messages to reduce the risk of users being scammed on social media. Include specific messages for when users are excited."

[0288] 2. "When a user receives a suspicious email, please generate a message that provides appropriate advice in case they feel uneasy."

[0289] The above is an embodiment of the digital lifeguard system according to the present invention. This system provides an environment in which users can safely engage in online activities and realizes risk management that takes into account their emotional state.

[0290] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0291] Step 1: User begins online activity

[0292] A user accesses the Internet using a device and begins searching, posting, and communicating. Specifically, the user opens a browser, types "cheapest brand name products" in the search bar, and clicks the search button. This causes the device to begin recording the user's actions (input).

[0293] Step 2: Capture behavioral and emotional data

[0294] The device captures the user's behavioral data (browser history, click patterns, keyboard input, etc.). At the same time, the device's camera and microphone are used to capture the user's video and audio data (input). Specifically, the device's camera records the user's facial expressions in real time, and the microphone collects the tone of voice. This data is then sent to the emotion engine (output).

[0295] Step 3: Sending behavioral and emotional data

[0296] The device encrypts the captured behavioral and emotional data (data processing) and transmits it to the server (output), which then receives the data for analysis.

[0297] Step 4: Risk analysis

[0298] The server inputs the received behavioral data into the generative model and evaluates the risk level (data calculation). For example, if a user attempts to click on a suspicious link, the generative model compares it with a database of past crimes and determines that it is a high risk. It then generates a risk level (output).

[0299] Step 5: Sentiment analysis

[0300] The server inputs the received emotion data into the emotion engine to evaluate the user's emotional state (data calculation). Specifically, the emotion engine analyzes facial expressions from the video data and evaluates the tone of voice from the audio data. This allows it to determine, for example, that the user is in a state of stress, and generates an emotional state (output).

[0301] Step 6: Create a warning based on risk and sentiment

[0302] The server generates an appropriate warning message based on the analyzed risk level and emotional state (data calculation). For example, if the user is feeling stressed, it creates a gentle warning message about the risk of fraud. It generates a message such as, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with authorized retailers when you have time." (output).

[0303] Step 7: Sending a warning message

[0304] The server sends the generated warning message to the user's terminal (output), which is then ready to display this warning in real time.

[0305] Step 8: Displaying warning messages and providing countermeasure information

[0306] The device displays the received warning message in real time and provides additional countermeasure information. For example, it may display on the screen, "If you fall victim to a scam, you are expected to lose 200,000 yen. Please contact the following organization for assistance," along with the estimated loss amount and information on where to seek assistance (output).

[0307] Step 9: Gather user reactions and feedback

[0308] The server records how users respond to warning messages and collects user reaction data (input). For example, it records the number of times users clicked links after receiving a warning, or the time it took them to close the page. This data is used as training data for the generative model and emotion engine, and is used to improve the system in future releases (data calculation and output).

[0309] (Application example 2)

[0310] 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."

[0311] In today's digital society, users are increasingly exposed to the risk of online crime and fraud. As a result, many people suffer financial losses and psychological stress. However, conventional security systems simply detect risks and do not consider whether the warning is appropriate for the user's emotional state, resulting in inappropriate timing and formatting of warnings. Therefore, there is a need for warnings that take into account the user's emotional state and deliver warnings in an appropriate tone.

[0312] 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.

[0313] In this invention, the server includes means for monitoring users' online behavior, means for analyzing users' behavioral data and assessing risk levels, means for displaying warnings for risky behavior, means including an emotion engine for analyzing users' emotional states, means for adjusting warning content based on the emotional states, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the AI ​​model. This makes it possible to not only detect online risks but also provide appropriate warnings and advice that take into account the users' emotional states.

[0314] "Online behavior monitoring means" refers to a function that records and tracks the activities that users perform on the Internet (searches, browsing, sending and receiving emails, etc.).

[0315] "Means for analyzing behavioral data" refers to a function for analyzing data collected from users' online behavior and assessing risk levels and trends.

[0316] The "warning display means" is a function for displaying a warning to the user about risky behavior based on the analysis results.

[0317] The "emotion engine" is a function that analyzes the user's emotional state in real time and uses that data to diagnose the risks of behavior and adjust warning content.

[0318] The "means for adjusting the warning content" is a function for appropriately changing the content and tone of the warning message based on the user's emotional state analyzed by the emotion engine.

[0319] "Means for providing information on estimated damages and consultation services" is a function for providing users with the estimated monetary damages due to risky behavior and information on appropriate consultation services.

[0320] The "user response recording means" is a function that records the user's response to the warning and uses that data to update the system's AI model and improve its accuracy.

[0321] The digital lifeguard system of the present invention is a system that mainly includes functions of monitoring users' online behavior, analyzing behavioral and emotional data, assessing risks, displaying warnings, providing estimated damage amounts and information on where to seek help, and recording users' responses. This system is designed to help users stay safe in the online environment.

[0322] System configuration

[0323] The system consists of the following main components:

[0324] 1. Server: This is the main component that receives and analyzes users' online behavioral data and emotional data. It is equipped with an emotion engine and generative AI model.

[0325] 2. Device: A device (e.g., PC, smartphone, tablet, etc.) that allows a user to access the Internet. The device is equipped with a camera and microphone to capture behavioral and emotional data.

[0326] 3. Generative AI model: An AI model that learns from past crime data and performs risk assessment based on user behavior data.

[0327] 4. Emotion Engine: This component analyzes the user's emotional state in real time and uses that data to adjust the content of warnings.

[0328] Hardware and Software Used

[0329] Hardware:

[0330] Device camera: Uses OpenCV for facial recognition and emotion analysis.

[0331] Device microphone: Used to analyze emotional state through voice analysis using Librosa.

[0332] software:

[0333] OpenCV: A library for capturing and processing video data.

[0334] Librosa: A library for capturing and analyzing audio data.

[0335] Keras: A framework for emotion recognition using deep learning models.

[0336] Joblib: A library for loading pre-trained generative models for risk assessment.

[0337] Requests: A library for communicating with the server.

[0338] Data processing and calculation

[0339] The server receives the user's online behavioral data and emotion data sent from the device, and processes the data in the following steps:

[0340] Behavioral Data Analysis: Generative AI models are used to assess a user's overall online behavior and calculate their risk level.

[0341] Emotional Data Analysis: The emotion engine analyzes the video and audio data captured from the camera and microphone to determine the user's emotional state.

[0342] Warning message generation: The server generates appropriate warning messages based on risk level and emotional state, adjusting their tone and content.

[0343] Provision of information: Provide users with estimated damage amounts and information on where to seek advice, and suggest specific countermeasures.

[0344] Specific examples

[0345] Example 1: If a user tries to access a fraudulent website, the system will detect the risk and, if it determines that the user is stressed, will display a gentle warning saying, "This site poses a risk of fraud. Do not click on this link."

[0346] Example prompt sentence:

[0347] "We want to develop an application that uses artificial intelligence to analyze emotions and warn users of risky behavior. We want to combine data from a user's online activity, camera, and microphone to perform real-time emotion analysis and risk assessment. We call this system a 'security assistant application,' and it will help users avoid becoming victims of fraud and crime."

[0348] Thus, the present invention provides a system that can appropriately manage the risks of online activities while taking into consideration the emotional state of the user.

[0349] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0350] Step 1: "User begins online activity"

[0351] A user accesses the Internet using a device and performs online activities such as searching, browsing, and sending and receiving emails. The input is the user's various online activities, which are captured by the device in real time. The output is the behavioral data captured by the device.

[0352] Step 2: Capture behavioral and emotional data

[0353] The device captures the user's online behavioral data and simultaneously captures emotional data in real time using video and audio data obtained from a camera and microphone. The input is the user's facial expression and audio data, which the device collects and sends to the server. The output is the behavioral and emotional data sent to the server.

[0354] Step 3: "Behavioral Data Risk Assessment"

[0355] The server inputs the behavioral data sent from the device into a generative AI model and evaluates the risk level based on past crime data. Specifically, the generative AI model analyzes the behavioral data and determines whether there is a risk of fraud or crime. The input is the behavioral data sent to the server, and the output is the risk level assessment result.

[0356] Step 4: "Analyzing Emotional Data"

[0357] The server analyzes the emotion data using an emotion engine to evaluate the user's emotional state. The emotion engine performs facial recognition from video data and estimates emotions from audio data. The input is the video and audio data sent to the server, and the output is the user's emotional state (e.g., stress, anxiety, relief, etc.).

[0358] Step 5: Generate a warning message

[0359] The server generates an appropriate warning message based on the risk level assessment results and emotional state assessment results. Specifically, if the risk level is high, the content and tone of the message are adjusted according to the user's emotional state. For example, if the user is feeling stressed, the warning is issued in a gentle tone. The input is the risk level assessment results and emotional state assessment results, and the output is a warning message.

[0360] Step 6: "Displaying warning messages and providing countermeasure information"

[0361] The terminal displays the warning message sent from the server in real time, along with the estimated loss amount and information on where to seek advice. Specifically, the terminal displays the message "There is a risk of fraud. Do not click on this link." It also provides information such as "If you fall victim to a fraud, you could lose 200,000 yen. Please contact the following organization." The input is the warning message sent from the server, and the output is the warning and countermeasure information displayed on the terminal.

[0362] Step 7: User Response and Feedback

[0363] The server records how the user responded to the warning (e.g., did not click on the link, ignored the warning). It also tracks changes in the user's emotional state and feeds this data back as learning data for the AI ​​model. The input is the user's reaction data and data on changes in emotional state, and the output is an updated AI model. This improves the accuracy of the entire system.

[0364] 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.

[0365] 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.

[0366] 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.

[0367] [Second embodiment]

[0368] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0369] 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.

[0370] 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).

[0371] 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.

[0372] 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.

[0373] 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).

[0374] 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.

[0375] 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.

[0376] 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.

[0377] 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.

[0378] 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.

[0379] 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."

[0380] The following describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive service for reducing the risk of users becoming involved in crimes or becoming victims online, and specifically operates in the following manner.

[0381] System Configuration

[0382] The system mainly consists of the following components:

[0383] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral data.

[0384] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[0385] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0386] Program processing and operation explanation

[0387] Users begin online activities

[0388] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[0389] Behavioral Data Capture

[0390] The server captures user behavior data in real time. To protect user privacy, it does not collect data that can identify personal information. User search queries, postings, received emails, etc. are monitored.

[0391] Risk Analysis

[0392] The captured behavioral data is input into a generative model stored on the server. The generative model learns from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if the search query "cheapest brand name products" is determined to be likely to be related to fraud, it will be classified as high risk.

[0393] Displaying warnings

[0394] If the risk is deemed high, the server will generate an appropriate warning message for the user, which will be displayed on the user's device. For example, the message may read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0395] Providing estimated damage amounts and information on where to seek advice

[0396] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0397] User Reactions and Feedback

[0398] The server records whether the user ignores the warning or whether they act on it, and the recorded data is used as training data for the generative model and fed back to improve the accuracy of the entire system.

[0399] Specific examples

[0400] Example 1: If a user posts on social media, "I want to buy expensive brand-name goods at a discount," the server will scan the content and detect risks. If applicable, it will display a warning and provide an estimated loss amount and information on where to seek advice.

[0401] Example 2: When a user receives an email containing a suspicious link, the server analyzes the email and displays a warning if it determines that the risk is high. It also notifies the user of the damage that could result from clicking the link and urges them to take action immediately.

[0402] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0403] The processing flow will be explained below.

[0404] Step 1:

[0405] The user begins their online journey. They use their device to access the Internet and search, write, and communicate. For example, they enter a search query such as "cheapest brand name products."

[0406] Step 2:

[0407] The device captures the user's online behavior, including search queries, social media posts, and received emails, and sends this data to a server.

[0408] Step 3:

[0409] The server receives the behavioral data sent from the device in real time, where it undergoes pre-processing and privacy-conscious data filtering.

[0410] Step 4:

[0411] The server inputs the received data into an internal generative model, which has learned from past crime data and fraud patterns to assess the risk level. For example, it analyzes search results for "cheapest brand name products" and determines that there is a high probability of fraud.

[0412] Step 5:

[0413] The server receives the evaluation results from the generative model and determines the risk level. It is classified as high risk, medium risk, or low risk. If there is a high possibility of fraud, it is classified as high risk.

[0414] Step 6:

[0415] If the server determines that the risk is high, it will generate an appropriate warning message for the user, which will include specific risks and countermeasures.

[0416] Step 7:

[0417] The server generates a warning message and sends it to the user's device, such as "This search result may contain fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0418] Step 8:

[0419] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[0420] Step 9:

[0421] The server generates a warning message along with estimated damages and information on where to seek advice, and provides it to the user. For example, it might say, "If you fall victim to a scam, you are likely to lose 200,000 yen. Please contact the following organization for advice."

[0422] Step 10:

[0423] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[0424] Step 11:

[0425] The server records how the user responded to the warning, for example, by storing a log of whether the warning was ignored, acted upon, or whether a helpline was contacted.

[0426] Step 12:

[0427] The server uses the recorded data as training data for the generative model, and the feedback improves the accuracy of the generative model, adjusting it to make future risk analyses more accurate.

[0428] Through the above process, the Digital Lifeguard System effectively reduces the risk of online crime and fraud, providing users with an environment where they can use the Internet with peace of mind.

[0429] Example 1

[0430] 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."

[0431] As the number of Internet users increases daily, the risk of becoming involved in online crimes and fraud is also increasing. In particular, many users suffer financial losses due to clicking on fraudulent links or visiting fraudulent websites. However, existing security measures have difficulty monitoring users' online behavior in real time and immediately warning them of risks, so effective countermeasures are needed.

[0432] 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.

[0433] In this invention, the server includes means for monitoring users' online behavior, means for capturing and preprocessing user behavior data in real time, means for inputting the captured data into a generative AI model and evaluating the risk level, means for displaying a warning for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the generative AI model. This provides an environment in which users can act safely online, and has the effect of significantly reducing the risk of crime and fraud.

[0434] "Online behavior" refers to a series of activities that users perform on the Internet, such as searching, posting, and receiving emails.

[0435] "Behavioral Data" is data that contains information about your online behavior, such as search queries, posts, and received emails.

[0436] "Real-time" means that data is processed as soon as it is generated, allowing for immediate analysis and evaluation without delay.

[0437] "Capture" is the act of collecting data about a user's online behavior.

[0438] "Preprocessing" is the process of removing unnecessary information from the captured raw data and converting it into a form suitable for analysis.

[0439] A "generative AI model" is an artificial intelligence algorithm that learns from past crime data and assesses risk based on new data.

[0440] A "risk level" is a number or category that evaluates the degree of danger associated with a user's behavior.

[0441] A "warning message" is a notification that alerts users to risky behavior.

[0442] "Estimated damage amount" is the amount of financial loss that a user is expected to incur if they fall victim to fraud or crime.

[0443] "Consultation information" is contact information for institutions or organizations that users should contact if they become victims.

[0444] "Recording a response" means recording how the user responded to the warning (whether they ignored it or followed it).

[0445] "Updating the AI ​​model" is the process of relearning the artificial intelligence algorithm based on recorded reaction data to improve its accuracy.

[0446] MODE FOR CARRYING OUT THE INVENTION

[0447] This invention is a system for reducing the risk of being involved in online crime and fraud by monitoring users' online behavior, assessing risks, and providing appropriate warning messages. The system mainly consists of three main components: a server, a user's device, and a generative AI model.

[0448] System Configuration

[0449] 1. Server:

[0450] The server captures and preprocesses users' online behavioral data in real time. It is preferable to use cloud services such as AWS (Amazon Web Services) or Google Cloud Platform. The server also contains a generative AI model built using machine learning libraries such as TensorFlow and PyTorch. This model evaluates risk levels based on the captured data.

[0451] 2. Terminal:

[0452] A terminal is a device that a user uses to access the internet, such as a PC, smartphone, or tablet. The terminal is responsible for displaying warning messages and consultation information sent from the server.

[0453] 3. Generative AI Model:

[0454] Generative AI models are trained using machine learning libraries such as TensorFlow and PyTorch based on historical crime data and fraud patterns, and then analyze captured user behavior data to assess risk levels.

[0455] Specific examples

[0456] Example 1: Risk assessment of search queries

[0457] When a user searches for "cheapest branded goods" on their device, the server captures the search query in real time. The captured data is input into a generative AI model, which assesses the risk based on past fraud patterns. If the risk is determined to be high, the server generates a warning message that reads, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers." and displays it on the device.

[0458] Example 2: Analysis of suspicious emails

[0459] When a user receives an email containing a suspicious link, the server captures the email in real time and performs text analysis. If the generative AI model determines that the email poses a high risk, the server generates a warning message that reads, "This email may be fraudulent. Do not click on the link." and displays it on the device.

[0460] Prompt Sentence Examples

[0461] "Analyze current user behavior based on past fraud data, conduct risk assessments, and generate warning messages if deemed high risk. Also consider countermeasures if users ignore the messages."

[0462] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0463] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0464] Step 1:

[0465] A user accesses the Internet. Using a device (PC, smartphone, tablet, etc.), the user opens a web browser and searches for, for example, "cheapest brand-name products."

[0466] Input: search query "cheapest brand name products".

[0467] Output: The user's search query is retrieved.

[0468] Step 2:

[0469] The server captures user behavior data in real time, including the search queries users type, the content they post, and the emails they receive.

[0470] Input: Your online behavior data (search queries, posts, received emails, etc.).

[0471] Output: Captured behavioral data.

[0472] Step 3:

[0473] The server preprocesses the captured behavioral data. This involves filtering unnecessary information from the text data and converting it into a format suitable for analysis. For example, natural language processing (NLP) techniques are used to extract important keywords.

[0474] Input: Raw captured data.

[0475] Output: Preprocessed data (data from which important keywords have been extracted).

[0476] Step 4:

[0477] The server then inputs the preprocessed data into a generative AI model, which uses historical crime data and fraud patterns to assess risk levels. For example, it might determine that a query like "cheapest brand name products" is likely to be fraud-related.

[0478] Input: Preprocessed data.

[0479] Output: Risk level assessment result (high risk, low risk, etc.).

[0480] Step 5:

[0481] If the server determines that the risk is high, it generates a warning message, which is then displayed on the user's device. For example, the message might read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0482] Input: Risk level assessment result (high risk).

[0483] Output: A warning message.

[0484] Step 6:

[0485] The server calculates the estimated loss if you fall victim to fraud or crime and provides it to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0486] Input: Risk level assessment result (high risk).

[0487] Output: Estimated damage amount and consultation information.

[0488] Step 7:

[0489] The server records whether the user ignored or acted on the warning. This record includes the user's actions after the warning (whether they clicked on a link, acted on the warning, etc.). The recorded data is used as training data for the generative AI model.

[0490] Input: User response data.

[0491] Output: An updated generative AI model.

[0492] In this way, the system monitors users' online behavior in real time and issues alerts when necessary, reducing the risk of crime and fraud.

[0493] (Application example 1)

[0494] 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."

[0495] In modern society, the number of internet-based frauds and crimes is on the rise, creating an urgent need to provide an environment where users can safely engage online. However, current security services lack real-time monitoring of online behavior and immediate warnings, limiting their effectiveness in preventing damage. Furthermore, they lack a feedback mechanism that allows the system to evolve based on user responses, leaving challenges in improving the overall accuracy of the system. To address these challenges, a comprehensive system with real-time monitoring, immediate warnings, and a feedback mechanism is needed.

[0496] 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.

[0497] In this invention, the server includes means for monitoring users' online behavior, means for analyzing user behavior data and assessing risk levels, means for displaying warnings for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, means for recording users' responses and updating the generative model, means for monitoring user behavior in real time, and means for analyzing risks in real time and displaying immediate warnings. This provides an environment where users can engage in online activities with peace of mind, making it possible to significantly reduce the risk of crime and fraud.

[0498] "User" means any person or entity that uses the System to conduct online activities.

[0499] "Online behavior" is a general term for actions such as searching, posting, and communication that are carried out using the Internet.

[0500] A "server" is a central component of the system and is an information processing device that collects, analyzes, and stores user behavior data.

[0501] "Behavioral Data" is a record of certain actions you take online, including search queries, postings, and the content of emails you receive and send.

[0502] "Risk level" is an indicator that shows the likelihood that a user's behavior will lead to crime or fraud.

[0503] A "warning" is a message that warns users against high-risk behavior.

[0504] "Estimated damage amount" is the amount of financial damage expected if a user falls victim to fraud or crime.

[0505] "Consultation information" refers to contact information for the police, consumer centers, etc. that users can consult if they are victims of a crime or fraud.

[0506] A "generative model" is an artificial intelligence model that learns from past crime data and is used to analyze the risks associated with user behavior.

[0507] A "feedback mechanism" is a mechanism for recording user responses and using that data to improve the accuracy of the generative model.

[0508] "Real-time monitoring" refers to the monitoring of users' online behavior in real time.

[0509] "Immediate warning" means that a warning message is displayed immediately when high-risk behavior is detected.

[0510] MODE FOR CARRYING OUT THE INVENTION

[0511] This paper describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive security service that reduces the risk of users becoming involved in online crimes or becoming victims. Below, each component of the system and its operation will be specifically described.

[0512] System Configuration

[0513] The system mainly consists of the following components:

[0514] 1. Server: This is the main component that monitors users' online behavior and analyzes the behavioral data.

[0515] 2. Terminal: A device (personal computer, smartphone, tablet, etc.) that a user uses to access the Internet.

[0516] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0517] Program processing and operation explanation

[0518] Monitoring your online behavior

[0519] Users access the Internet using their devices to search, post, and communicate. For example, suppose a user searches for "cheapest brand-name products." All of this behavior is monitored in real time.

[0520] Behavioral data capture and analysis

[0521] The server captures user behavior data in real time and inputs this data into a generative model. The generative model has learned from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if a search query such as "cheapest brand name products" is determined to be highly likely to be related to fraud, it will be classified as high risk.

[0522] Risk analysis and warning display

[0523] If the risk is determined to be high in real time, the server generates an appropriate warning message for the user and displays it on the device, such as "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from an authorized retailer."

[0524] Providing estimated damage amounts and information on where to seek advice

[0525] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0526] User Reactions and Feedback

[0527] The server records whether the user ignores the warning or whether they act on it. The recorded data is used as training data for the generative model and as feedback to improve the accuracy of the entire system.

[0528] Hardware and software used

[0529] Hardware: Personal computers, smartphones, tablets

[0530] software:

[0531] AI Library: Used to build generative AI models and perform risk analysis

[0532] Monitoring Tools: Monitor your users' online behavior with UserActionMonitor

[0533] Notification Management System: Display alerts using NotificationManager

[0534] Examples and prompts

[0535] Specific examples

[0536] If a user posts on social media that they want to buy expensive branded goods at a discount, the server will scan the post and detect any risks. If applicable, it will display a warning, provide an estimated loss amount, and provide information on where to seek advice.

[0537] Prompt Sentence Examples

[0538] "When a user searches for 'cheapest branded goods,' please generate a prompt to check for fraud risk."

[0539] As described above, the digital lifeguard system according to the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0540] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0541] Step 1:

[0542] A user accesses the Internet using a device. This initiates online activities such as search queries, posting, and receiving / sending emails. The user's online behavior data (search queries, social media postings, email content) is obtained as input. This becomes the initial input for the system.

[0543] Step 2:

[0544] The server monitors the user's online behavior in real time. Here, we use the UserActionMonitor tool to capture the user's behavior data. As input, the user's online behavior data is monitored and sent to the server as behavior data. As output, this data is saved as behavior data capture.

[0545] Step 3:

[0546] The server inputs the captured behavioral data into a generative AI model. This AI model has learned from past crime data and uses an algorithm to assess risk levels. The captured behavioral data is passed to the generative AI model as input. The output is an assessment of the risk level for the behavioral data.

[0547] Step 4:

[0548] The server makes a decision based on the risk level assessed by the generative AI model. A warning message is generated for actions with a high risk level. Here, the WarningGenerator tool is used. The input is the risk level assessment result. The output is a generated warning message.

[0549] Step 5:

[0550] The server notifies the user's terminal of the generated warning message. Here, we use the NotificationManager tool. As input, there is the generated warning message, which is sent to the user's terminal. As output, the warning message is displayed on the user's terminal.

[0551] Step 6:

[0552] The server calculates and provides estimated damage amounts and information on where to seek advice for high-risk behavior. The inputs are behavioral data and risk level assessment results. The output is a calculated damage amount and information on where to seek advice, which are provided to the user.

[0553] Step 7:

[0554] It monitors and records how users react to warnings and provided information. The server collects this reaction data and uses it as training data for the generative model. As input, user reaction data is collected. As output, this data is saved as feedback for the generative model.

[0555] These steps enable the Digital Lifeguard system to monitor users' online behavior in real time and provide appropriate warnings and information.

[0556] 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.

[0557] This paper describes a digital lifeguard system incorporating an emotion engine, which is an interactive service for users to reduce the risk of becoming involved in online crimes and victimization. The system also recognizes the user's emotional state and provides warnings and advice based on that state.

[0558] System Configuration

[0559] The system mainly consists of the following components:

[0560] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral and emotional data.

[0561] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[0562] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0563] 4. Emotion Engine: A component that recognizes and analyzes user emotions in real time.

[0564] Program processing and operation explanation

[0565] Users begin online activities

[0566] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[0567] Capturing behavioral and emotional data

[0568] The device captures the user's online behavior and input from the camera and microphone. The emotion engine analyzes the user's emotions using the captured video and audio data. This data is sent to the server.

[0569] Risk and Sentiment Analysis

[0570] The server receives and analyzes the behavioral and emotional data. The behavioral data is input into a generative model to assess the risk level. The emotional data is analyzed by an emotion engine to assess the user's current emotional state (e.g., stress, relief, anxiety).

[0571] Generating Emotion-Based Alerts

[0572] The server generates an appropriate warning message based on the risk level and the evaluation of the user's emotional state. For example, if the user is feeling stressed, the server can issue a warning in a gentle tone that takes into account the user's psychological state. A specific example of a message that could be generated is, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time."

[0573] Displaying warnings and providing countermeasure information

[0574] The server sends the generated warning message to the user's device. The device displays the warning message in real time, along with the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization for advice."

[0575] User Reactions and Feedback

[0576] The server records how users respond to the alerts and how their emotional state changes. These data are used as training data for the generative model and emotion engine, and are fed back to improve the accuracy of the entire system.

[0577] Specific examples

[0578] Example 1: If a user posts on social media, "I want to buy expensive branded goods at a discount," the server will scan the content and the user's emotions to detect the risk of fraud. Also, if the user's emotional state is "excited," the server will display a warning in a gentle tone to help them regain their composure.

[0579] Example 2: If a user receives an email containing a suspicious link, the server analyzes the email and, if it detects that the user's emotional state is "anxiety," it suggests ways to mitigate the risk and provides a message to help the user stay calm.

[0580] In this way, the digital lifeguard system of the present invention, which is combined with an emotion engine, more effectively provides an environment in which users can safely engage online, and realizes risk management that takes into account their emotional state.

[0581] The processing flow will be explained below.

[0582] Step 1:

[0583] The user begins their online journey. They use their device to access the Internet and search, post, and communicate. For example, they search for "cheapest brand name products."

[0584] Step 2:

[0585] The device captures the user's online behavior, such as search queries, social media posts, and received emails, and sends this data to a server.

[0586] Step 3:

[0587] The device's camera and microphone are used to capture emotional data from the user's facial expressions and voice, and this data is also sent to the server.

[0588] Step 4:

[0589] The server receives behavioral and emotional data sent from the device in real time, and performs privacy-conscious data filtering at this stage.

[0590] Step 5:

[0591] The server inputs the received behavioral data into a generative model, which is trained on historical crime data and fraud patterns and uses this data to assess risk levels. For example, a search query for "cheapest branded goods" might be deemed to have a high risk of fraud.

[0592] Step 6:

[0593] The server analyzes the emotion data using an emotion engine, evaluating the user's emotional state based on their facial expressions and voice, and detecting, for example, "anxiety" or "excitement."

[0594] Step 7:

[0595] The server combines the risk level assessment results with the emotional state assessment results to generate an appropriate warning message. If the risk is high and the user is feeling "anxious," it creates a gentler warning that takes into consideration the user's feelings. For example, it might say, "This search result may include fraudulent sites selling counterfeit branded goods. Please check with authorized retailers when you have time."

[0596] Step 8:

[0597] The server sends the generated warning message to the user's device. At the same time, it also generates an estimated loss amount and information on where to seek advice. For example, "If you fall victim to a scam, you can expect to lose 200,000 yen. Please contact the following organizations for advice."

[0598] Step 9:

[0599] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[0600] Step 10:

[0601] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[0602] Step 11:

[0603] The server records how the user responded to the warning, for example, whether the warning was ignored or acted upon, and also records the user's emotional state in response to the warning.

[0604] Step 12:

[0605] The server uses the recorded behavioral and emotional data as training data for the generative model and emotion engine, improving the accuracy of the entire system and adjusting it to ensure more accurate risk and emotion analysis from the next time onwards.

[0606] Through the above process, by combining an emotion engine with the digital lifeguard system, user safety can be ensured more effectively.

[0607] Example 2

[0608] 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."

[0609] In today's digital society, users are at a greater risk of becoming involved in crimes and fraud online. Furthermore, warnings and advice that ignore users' emotional state make it difficult to encourage appropriate responses. As a result, users' safety may not be adequately ensured, and the damage may worsen.

[0610] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing user behavioral data and emotional data and transmitting them to the server, means for inputting the user behavioral data into a generative model to evaluate the risk level, and means for analyzing the emotional data using an emotion engine to evaluate the user's emotional state. This ensures the safety of the user and enables appropriate warnings and advice that take into account the user's emotional state.

[0611] "Means for monitoring users' online behavior" refers to the ability to track and record users' activities on the Internet, such as searches, posts, and clicks, in real time.

[0612] "Means for capturing user behavioral data and emotional data and transmitting it to the server" refers to a function that collects the user's operation history, facial expressions, tone of voice, etc., and transfers this data to the server.

[0613] The "means of inputting into a generative model to assess risk levels" refers to a function that inputs acquired behavioral data into an AI model and performs risk assessment based on past crime data.

[0614] "Means for analyzing emotional data using an emotion engine and assessing the user's emotional state" is a function that analyzes captured video and audio data and determines the user's emotional state.

[0615] "Means for generating an appropriate warning message based on the risk level and emotional state and sending it to the user's device" refers to a function that creates an appropriate warning message for the user based on the risk assessment and emotional state, and immediately sends it to the user's device.

[0616] "Means for providing users with estimated damage amounts and information on consultation centers" is a function that provides users with information on appropriate consultation centers along with estimated damage amounts when a risk such as fraud is detected.

[0617] "Means for recording user responses and updating the AI ​​model" refers to a function that records user responses to warning messages and uses that data to improve the accuracy and efficiency of the AI ​​model.

[0618] A "generative model" is an artificial intelligence model that learns from past crime data and compares it with current data to perform risk assessment.

[0619] An "emotion engine" is a system element that analyzes emotional data such as images and sounds and determines the user's emotional state.

[0620] The digital lifeguard system of this invention ensures the safety of users online and provides appropriate warnings and advice taking into account their emotional state. This system mainly consists of four components: a server, a terminal, a generative model, and an emotion engine.

[0621] The server is the central part of the system and plays a key role in analyzing users' behavioral and emotional data. The terminal is a device (such as a PC, smartphone, or tablet) that users use to access the Internet, and is responsible for capturing behavioral and emotional data and sending it to the server. The generative model is installed in the server and is an AI model used to perform risk analysis based on past crime data. The emotion engine is a component that recognizes and analyzes users' emotions in real time.

[0622] The system begins operation when a user accesses the Internet using a device and begins searching, posting, or communicating. The device captures the user's online behavior and simultaneously captures input data from a camera and microphone. For example, the device's camera captures the user's facial expressions and the microphone records the user's tone of voice. This data is analyzed by an emotion engine to evaluate the user's emotional state.

[0623] The server receives the behavioral and emotional data sent from the device and uses it to analyze each. The behavioral data is input into a generative model to evaluate the risk level. For example, if a user attempts to click on a suspicious link, it is compared with a past database to determine whether the risk is high. The emotional data is analyzed by an emotion engine to evaluate the user's current emotional state. For example, it analyzes whether the user is feeling stressed while searching.

[0624] The server generates an appropriate warning message based on the risk level and the user's emotional state. If the user is feeling stressed, the server can issue a gentle warning that takes into account the user's psychological state. For example, a message such as "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time" can be generated.

[0625] The server sends the generated warning message to the user's device, which displays it in real time. It also displays the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization."

[0626] The server records how the user responded to the warning and also records how their emotional state changed. This data is used as training data for the generative model and emotion engine, and is fed back to improve the accuracy of the entire system.

[0627] Examples of prompt statements

[0628] Examples of prompts include:

[0629] 1. "Generate warning messages to reduce the risk of users being scammed on social media. Include specific messages for when users are excited."

[0630] 2. "When a user receives a suspicious email, please generate a message that provides appropriate advice in case they feel uneasy."

[0631] The above is an embodiment of the digital lifeguard system according to the present invention. This system provides an environment in which users can safely engage in online activities and realizes risk management that takes into account their emotional state.

[0632] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0633] Step 1: User begins online activity

[0634] A user accesses the Internet using a device and begins searching, posting, and communicating. Specifically, the user opens a browser, types "cheapest brand name products" in the search bar, and clicks the search button. This causes the device to begin recording the user's actions (input).

[0635] Step 2: Capture behavioral and emotional data

[0636] The device captures the user's behavioral data (browser history, click patterns, keyboard input, etc.). At the same time, the device's camera and microphone are used to capture the user's video and audio data (input). Specifically, the device's camera records the user's facial expressions in real time, and the microphone collects the tone of voice. This data is then sent to the emotion engine (output).

[0637] Step 3: Sending behavioral and emotional data

[0638] The device encrypts the captured behavioral and emotional data (data processing) and transmits it to the server (output), which then receives the data for analysis.

[0639] Step 4: Risk analysis

[0640] The server inputs the received behavioral data into the generative model and evaluates the risk level (data calculation). For example, if a user attempts to click on a suspicious link, the generative model compares it with a database of past crimes and determines that it is a high risk. It then generates a risk level (output).

[0641] Step 5: Sentiment analysis

[0642] The server inputs the received emotion data into the emotion engine to evaluate the user's emotional state (data calculation). Specifically, the emotion engine analyzes facial expressions from the video data and evaluates the tone of voice from the audio data. This allows it to determine, for example, that the user is in a state of stress, and generates an emotional state (output).

[0643] Step 6: Create a warning based on risk and sentiment

[0644] The server generates an appropriate warning message based on the analyzed risk level and emotional state (data calculation). For example, if the user is feeling stressed, it creates a gentle warning message about the risk of fraud. It generates a message such as, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with authorized retailers when you have time." (output).

[0645] Step 7: Sending a warning message

[0646] The server sends the generated warning message to the user's terminal (output), which is then ready to display this warning in real time.

[0647] Step 8: Displaying warning messages and providing countermeasure information

[0648] The device displays the received warning message in real time and provides additional countermeasure information. For example, it may display on the screen, "If you fall victim to a scam, you are expected to lose 200,000 yen. Please contact the following organization for assistance," along with the estimated loss amount and information on where to seek assistance (output).

[0649] Step 9: Gather user reactions and feedback

[0650] The server records how users respond to warning messages and collects user reaction data (input). For example, it records the number of times users clicked links after receiving a warning, or the time it took them to close the page. This data is used as training data for the generative model and emotion engine, and is used to improve the system in future releases (data calculation and output).

[0651] (Application example 2)

[0652] 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."

[0653] In today's digital society, users are increasingly exposed to the risk of online crime and fraud. As a result, many people suffer financial losses and psychological stress. However, conventional security systems simply detect risks and do not consider whether the warning is appropriate for the user's emotional state, resulting in inappropriate timing and formatting of warnings. Therefore, there is a need for warnings that take into account the user's emotional state and deliver warnings in an appropriate tone.

[0654] 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.

[0655] In this invention, the server includes means for monitoring users' online behavior, means for analyzing users' behavioral data and assessing risk levels, means for displaying warnings for risky behavior, means including an emotion engine for analyzing users' emotional states, means for adjusting warning content based on the emotional states, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the AI ​​model. This makes it possible to not only detect online risks but also provide appropriate warnings and advice that take into account the users' emotional states.

[0656] "Online behavior monitoring means" refers to a function that records and tracks the activities that users perform on the Internet (searches, browsing, sending and receiving emails, etc.).

[0657] "Means for analyzing behavioral data" refers to a function for analyzing data collected from users' online behavior and assessing risk levels and trends.

[0658] The "warning display means" is a function for displaying a warning to the user about risky behavior based on the analysis results.

[0659] The "emotion engine" is a function that analyzes the user's emotional state in real time and uses that data to diagnose the risks of behavior and adjust warning content.

[0660] The "means for adjusting the warning content" is a function for appropriately changing the content and tone of the warning message based on the user's emotional state analyzed by the emotion engine.

[0661] "Means for providing information on estimated damages and consultation services" is a function for providing users with the estimated monetary damages due to risky behavior and information on appropriate consultation services.

[0662] The "user response recording means" is a function that records the user's response to the warning and uses that data to update the system's AI model and improve its accuracy.

[0663] The digital lifeguard system of the present invention is a system that mainly includes functions of monitoring users' online behavior, analyzing behavioral and emotional data, assessing risks, displaying warnings, providing estimated damage amounts and information on where to seek help, and recording users' responses. This system is designed to help users stay safe in the online environment.

[0664] System configuration

[0665] The system consists of the following main components:

[0666] 1. Server: This is the main component that receives and analyzes users' online behavioral data and emotional data. It is equipped with an emotion engine and generative AI model.

[0667] 2. Device: A device (e.g., PC, smartphone, tablet, etc.) that allows a user to access the Internet. The device is equipped with a camera and microphone to capture behavioral and emotional data.

[0668] 3. Generative AI model: An AI model that learns from past crime data and performs risk assessment based on user behavior data.

[0669] 4. Emotion Engine: This component analyzes the user's emotional state in real time and uses that data to adjust the content of warnings.

[0670] Hardware and Software Used

[0671] Hardware:

[0672] Device camera: Uses OpenCV for facial recognition and emotion analysis.

[0673] Device microphone: Used to analyze emotional state through voice analysis using Librosa.

[0674] software:

[0675] OpenCV: A library for capturing and processing video data.

[0676] Librosa: A library for capturing and analyzing audio data.

[0677] Keras: A framework for emotion recognition using deep learning models.

[0678] Joblib: A library for loading pre-trained generative models for risk assessment.

[0679] Requests: A library for communicating with the server.

[0680] Data processing and calculation

[0681] The server receives the user's online behavioral data and emotion data sent from the device, and processes the data in the following steps:

[0682] Behavioral Data Analysis: Generative AI models are used to assess a user's overall online behavior and calculate their risk level.

[0683] Emotional Data Analysis: The emotion engine analyzes the video and audio data captured from the camera and microphone to determine the user's emotional state.

[0684] Warning message generation: The server generates appropriate warning messages based on risk level and emotional state, adjusting their tone and content.

[0685] Provision of information: Provide users with estimated damage amounts and information on where to seek advice, and suggest specific countermeasures.

[0686] Specific examples

[0687] Example 1: If a user tries to access a fraudulent website, the system will detect the risk and, if it determines that the user is stressed, will display a gentle warning saying, "This site poses a risk of fraud. Do not click on this link."

[0688] Example prompt sentence:

[0689] "We want to develop an application that uses artificial intelligence to analyze emotions and warn users of risky behavior. We want to combine data from a user's online activity, camera, and microphone to perform real-time emotion analysis and risk assessment. We call this system a 'security assistant application,' and it will help users avoid becoming victims of fraud and crime."

[0690] Thus, the present invention provides a system that can appropriately manage the risks of online activities while taking into consideration the emotional state of the user.

[0691] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0692] Step 1: "User begins online activity"

[0693] A user accesses the Internet using a device and performs online activities such as searching, browsing, and sending and receiving emails. The input is the user's various online activities, which are captured by the device in real time. The output is the behavioral data captured by the device.

[0694] Step 2: Capture behavioral and emotional data

[0695] The device captures the user's online behavioral data and simultaneously captures emotional data in real time using video and audio data obtained from a camera and microphone. The input is the user's facial expression and audio data, which the device collects and sends to the server. The output is the behavioral and emotional data sent to the server.

[0696] Step 3: "Behavioral Data Risk Assessment"

[0697] The server inputs the behavioral data sent from the device into a generative AI model and evaluates the risk level based on past crime data. Specifically, the generative AI model analyzes the behavioral data and determines whether there is a risk of fraud or crime. The input is the behavioral data sent to the server, and the output is the risk level assessment result.

[0698] Step 4: "Analyzing Emotional Data"

[0699] The server analyzes the emotion data using an emotion engine to evaluate the user's emotional state. The emotion engine performs facial recognition from video data and estimates emotions from audio data. The input is the video and audio data sent to the server, and the output is the user's emotional state (e.g., stress, anxiety, relief, etc.).

[0700] Step 5: Generate a warning message

[0701] The server generates an appropriate warning message based on the risk level assessment results and emotional state assessment results. Specifically, if the risk level is high, the content and tone of the message are adjusted according to the user's emotional state. For example, if the user is feeling stressed, the warning is issued in a gentle tone. The input is the risk level assessment results and emotional state assessment results, and the output is a warning message.

[0702] Step 6: "Displaying warning messages and providing countermeasure information"

[0703] The terminal displays the warning message sent from the server in real time, along with the estimated loss amount and information on where to seek advice. Specifically, the terminal displays the message "There is a risk of fraud. Do not click on this link." It also provides information such as "If you fall victim to a fraud, you could lose 200,000 yen. Please contact the following organization." The input is the warning message sent from the server, and the output is the warning and countermeasure information displayed on the terminal.

[0704] Step 7: User Response and Feedback

[0705] The server records how the user responded to the warning (e.g., did not click on the link, ignored the warning). It also tracks changes in the user's emotional state and feeds this data back as learning data for the AI ​​model. The input is the user's reaction data and data on changes in emotional state, and the output is an updated AI model. This improves the accuracy of the entire system.

[0706] 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.

[0707] 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.

[0708] 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.

[0709] [Third embodiment]

[0710] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0711] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0712] 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).

[0713] 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.

[0714] 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.

[0715] 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).

[0716] 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.

[0717] 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.

[0718] 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.

[0719] 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.

[0720] 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.

[0721] 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."

[0722] The following describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive service for reducing the risk of users becoming involved in crimes or becoming victims online, and specifically operates in the following manner.

[0723] System Configuration

[0724] The system mainly consists of the following components:

[0725] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral data.

[0726] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[0727] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0728] Program processing and operation explanation

[0729] Users begin online activities

[0730] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[0731] Behavioral Data Capture

[0732] The server captures user behavior data in real time. To protect user privacy, it does not collect data that can identify personal information. User search queries, postings, received emails, etc. are monitored.

[0733] Risk Analysis

[0734] The captured behavioral data is input into a generative model stored on the server. The generative model learns from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if the search query "cheapest brand name products" is determined to be likely to be related to fraud, it will be classified as high risk.

[0735] Displaying warnings

[0736] If the risk is deemed high, the server will generate an appropriate warning message for the user, which will be displayed on the user's device. For example, the message may read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0737] Providing estimated damage amounts and information on where to seek advice

[0738] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0739] User Reactions and Feedback

[0740] The server records whether the user ignores the warning or whether they act on it, and the recorded data is used as training data for the generative model and fed back to improve the accuracy of the entire system.

[0741] Specific examples

[0742] Example 1: If a user posts on social media, "I want to buy expensive brand-name goods at a discount," the server will scan the content and detect risks. If applicable, it will display a warning and provide an estimated loss amount and information on where to seek advice.

[0743] Example 2: When a user receives an email containing a suspicious link, the server analyzes the email and displays a warning if it determines that the risk is high. It also notifies the user of the damage that could result from clicking the link and urges them to take action immediately.

[0744] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0745] The processing flow will be explained below.

[0746] Step 1:

[0747] The user begins their online journey. They use their device to access the Internet and search, write, and communicate. For example, they enter a search query such as "cheapest brand name products."

[0748] Step 2:

[0749] The device captures the user's online behavior, including search queries, social media posts, and received emails, and sends this data to a server.

[0750] Step 3:

[0751] The server receives the behavioral data sent from the device in real time, where it undergoes pre-processing and privacy-conscious data filtering.

[0752] Step 4:

[0753] The server inputs the received data into an internal generative model, which has learned from past crime data and fraud patterns to assess the risk level. For example, it analyzes search results for "cheapest brand name products" and determines that there is a high probability of fraud.

[0754] Step 5:

[0755] The server receives the evaluation results from the generative model and determines the risk level. It is classified as high risk, medium risk, or low risk. If there is a high possibility of fraud, it is classified as high risk.

[0756] Step 6:

[0757] If the server determines that the risk is high, it will generate an appropriate warning message for the user, which will include specific risks and countermeasures.

[0758] Step 7:

[0759] The server generates a warning message and sends it to the user's device, such as "This search result may contain fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0760] Step 8:

[0761] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[0762] Step 9:

[0763] The server generates a warning message along with estimated damages and information on where to seek advice, and provides it to the user. For example, it might say, "If you fall victim to a scam, you are likely to lose 200,000 yen. Please contact the following organization for advice."

[0764] Step 10:

[0765] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[0766] Step 11:

[0767] The server records how the user responded to the warning, for example, by storing a log of whether the warning was ignored, acted upon, or whether a helpline was contacted.

[0768] Step 12:

[0769] The server uses the recorded data as training data for the generative model, and the feedback improves the accuracy of the generative model, adjusting it to make future risk analyses more accurate.

[0770] Through the above process, the Digital Lifeguard System effectively reduces the risk of online crime and fraud, providing users with an environment where they can use the Internet with peace of mind.

[0771] Example 1

[0772] 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."

[0773] As the number of Internet users increases daily, the risk of becoming involved in online crimes and fraud is also increasing. In particular, many users suffer financial losses due to clicking on fraudulent links or visiting fraudulent websites. However, existing security measures have difficulty monitoring users' online behavior in real time and immediately warning them of risks, so effective countermeasures are needed.

[0774] 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.

[0775] In this invention, the server includes means for monitoring users' online behavior, means for capturing and preprocessing user behavior data in real time, means for inputting the captured data into a generative AI model and evaluating the risk level, means for displaying a warning for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the generative AI model. This provides an environment in which users can act safely online, and has the effect of significantly reducing the risk of crime and fraud.

[0776] "Online behavior" refers to a series of activities that users perform on the Internet, such as searching, posting, and receiving emails.

[0777] "Behavioral Data" is data that contains information about your online behavior, such as search queries, posts, and received emails.

[0778] "Real-time" means that data is processed as soon as it is generated, allowing for immediate analysis and evaluation without delay.

[0779] "Capture" is the act of collecting data about a user's online behavior.

[0780] "Preprocessing" is the process of removing unnecessary information from the captured raw data and converting it into a form suitable for analysis.

[0781] A "generative AI model" is an artificial intelligence algorithm that learns from past crime data and assesses risk based on new data.

[0782] A "risk level" is a number or category that evaluates the degree of danger associated with a user's behavior.

[0783] A "warning message" is a notification that alerts users to risky behavior.

[0784] "Estimated damage amount" is the amount of financial loss that a user is expected to incur if they fall victim to fraud or crime.

[0785] "Consultation information" is contact information for institutions or organizations that users should contact if they become victims.

[0786] "Recording a response" means recording how the user responded to the warning (whether they ignored it or followed it).

[0787] "Updating the AI ​​model" is the process of relearning the artificial intelligence algorithm based on recorded reaction data to improve its accuracy.

[0788] MODE FOR CARRYING OUT THE INVENTION

[0789] This invention is a system for reducing the risk of being involved in online crime and fraud by monitoring users' online behavior, assessing risks, and providing appropriate warning messages. The system mainly consists of three main components: a server, a user's device, and a generative AI model.

[0790] System Configuration

[0791] 1. Server:

[0792] The server captures and preprocesses users' online behavioral data in real time. It is preferable to use cloud services such as AWS (Amazon Web Services) or Google Cloud Platform. The server also contains a generative AI model built using machine learning libraries such as TensorFlow and PyTorch. This model evaluates risk levels based on the captured data.

[0793] 2. Terminal:

[0794] A terminal is a device that a user uses to access the internet, such as a PC, smartphone, or tablet. The terminal is responsible for displaying warning messages and consultation information sent from the server.

[0795] 3. Generative AI Model:

[0796] Generative AI models are trained using machine learning libraries such as TensorFlow and PyTorch based on historical crime data and fraud patterns, and then analyze captured user behavior data to assess risk levels.

[0797] Specific examples

[0798] Example 1: Risk assessment of search queries

[0799] When a user searches for "cheapest branded goods" on their device, the server captures the search query in real time. The captured data is input into a generative AI model, which assesses the risk based on past fraud patterns. If the risk is determined to be high, the server generates a warning message that reads, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers." and displays it on the device.

[0800] Example 2: Analysis of suspicious emails

[0801] When a user receives an email containing a suspicious link, the server captures the email in real time and performs text analysis. If the generative AI model determines that the email poses a high risk, the server generates a warning message that reads, "This email may be fraudulent. Do not click on the link." and displays it on the device.

[0802] Prompt Sentence Examples

[0803] "Analyze current user behavior based on past fraud data, conduct risk assessments, and generate warning messages if deemed high risk. Also consider countermeasures if users ignore the messages."

[0804] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

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

[0806] Step 1:

[0807] A user accesses the Internet. Using a device (PC, smartphone, tablet, etc.), the user opens a web browser and searches for, for example, "cheapest brand-name products."

[0808] Input: search query "cheapest brand name products".

[0809] Output: The user's search query is retrieved.

[0810] Step 2:

[0811] The server captures user behavior data in real time, including the search queries users type, the content they post, and the emails they receive.

[0812] Input: Your online behavior data (search queries, posts, received emails, etc.).

[0813] Output: Captured behavioral data.

[0814] Step 3:

[0815] The server preprocesses the captured behavioral data. This involves filtering unnecessary information from the text data and converting it into a format suitable for analysis. For example, natural language processing (NLP) techniques are used to extract important keywords.

[0816] Input: Raw captured data.

[0817] Output: Preprocessed data (data from which important keywords have been extracted).

[0818] Step 4:

[0819] The server then inputs the preprocessed data into a generative AI model, which uses historical crime data and fraud patterns to assess risk levels. For example, it might determine that a query like "cheapest brand name products" is likely to be fraud-related.

[0820] Input: Preprocessed data.

[0821] Output: Risk level assessment result (high risk, low risk, etc.).

[0822] Step 5:

[0823] If the server determines that the risk is high, it generates a warning message, which is then displayed on the user's device. For example, the message might read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[0824] Input: Risk level assessment result (high risk).

[0825] Output: A warning message.

[0826] Step 6:

[0827] The server calculates the estimated loss if you fall victim to fraud or crime and provides it to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0828] Input: Risk level assessment result (high risk).

[0829] Output: Estimated damage amount and consultation information.

[0830] Step 7:

[0831] The server records whether the user ignored or acted on the warning. This record includes the user's actions after the warning (whether they clicked on a link, acted on the warning, etc.). The recorded data is used as training data for the generative AI model.

[0832] Input: User response data.

[0833] Output: An updated generative AI model.

[0834] In this way, the system monitors users' online behavior in real time and issues alerts when necessary, reducing the risk of crime and fraud.

[0835] (Application example 1)

[0836] 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."

[0837] In modern society, the number of internet-based frauds and crimes is on the rise, creating an urgent need to provide an environment where users can safely engage online. However, current security services lack real-time monitoring of online behavior and immediate warnings, limiting their effectiveness in preventing damage. Furthermore, they lack a feedback mechanism that allows the system to evolve based on user responses, leaving challenges in improving the overall accuracy of the system. To address these challenges, a comprehensive system with real-time monitoring, immediate warnings, and a feedback mechanism is needed.

[0838] 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.

[0839] In this invention, the server includes means for monitoring users' online behavior, means for analyzing user behavior data and assessing risk levels, means for displaying warnings for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, means for recording users' responses and updating the generative model, means for monitoring user behavior in real time, and means for analyzing risks in real time and displaying immediate warnings. This provides an environment where users can engage in online activities with peace of mind, making it possible to significantly reduce the risk of crime and fraud.

[0840] "User" means any person or entity that uses the System to conduct online activities.

[0841] "Online behavior" is a general term for actions such as searching, posting, and communication that are carried out using the Internet.

[0842] A "server" is a central component of the system and is an information processing device that collects, analyzes, and stores user behavior data.

[0843] "Behavioral Data" is a record of certain actions you take online, including search queries, postings, and the content of emails you receive and send.

[0844] "Risk level" is an indicator that shows the likelihood that a user's behavior will lead to crime or fraud.

[0845] A "warning" is a message that warns users against high-risk behavior.

[0846] "Estimated damage amount" is the amount of financial damage expected if a user falls victim to fraud or crime.

[0847] "Consultation information" refers to contact information for the police, consumer centers, etc. that users can consult if they are victims of a crime or fraud.

[0848] A "generative model" is an artificial intelligence model that learns from past crime data and is used to analyze the risks associated with user behavior.

[0849] A "feedback mechanism" is a mechanism for recording user responses and using that data to improve the accuracy of the generative model.

[0850] "Real-time monitoring" refers to the monitoring of users' online behavior in real time.

[0851] "Immediate warning" means that a warning message is displayed immediately when high-risk behavior is detected.

[0852] MODE FOR CARRYING OUT THE INVENTION

[0853] This paper describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive security service that reduces the risk of users becoming involved in online crimes or becoming victims. Below, each component of the system and its operation will be specifically described.

[0854] System Configuration

[0855] The system mainly consists of the following components:

[0856] 1. Server: This is the main component that monitors users' online behavior and analyzes the behavioral data.

[0857] 2. Terminal: A device (personal computer, smartphone, tablet, etc.) that a user uses to access the Internet.

[0858] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0859] Program processing and operation explanation

[0860] Monitoring your online behavior

[0861] Users access the Internet using their devices to search, post, and communicate. For example, suppose a user searches for "cheapest brand-name products." All of this behavior is monitored in real time.

[0862] Behavioral data capture and analysis

[0863] The server captures user behavior data in real time and inputs this data into a generative model. The generative model has learned from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if a search query such as "cheapest brand name products" is determined to be highly likely to be related to fraud, it will be classified as high risk.

[0864] Risk analysis and warning display

[0865] If the risk is determined to be high in real time, the server generates an appropriate warning message for the user and displays it on the device, such as "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from an authorized retailer."

[0866] Providing estimated damage amounts and information on where to seek advice

[0867] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[0868] User Reactions and Feedback

[0869] The server records whether the user ignores the warning or whether they act on it. The recorded data is used as training data for the generative model and as feedback to improve the accuracy of the entire system.

[0870] Hardware and software used

[0871] Hardware: Personal computers, smartphones, tablets

[0872] software:

[0873] AI Library: Used to build generative AI models and perform risk analysis

[0874] Monitoring Tools: Monitor your users' online behavior with UserActionMonitor

[0875] Notification Management System: Display alerts using NotificationManager

[0876] Examples and prompts

[0877] Specific examples

[0878] If a user posts on social media that they want to buy expensive branded goods at a discount, the server will scan the post and detect any risks. If applicable, it will display a warning, provide an estimated loss amount, and provide information on where to seek advice.

[0879] Prompt Sentence Examples

[0880] "When a user searches for 'cheapest branded goods,' please generate a prompt to check for fraud risk."

[0881] As described above, the digital lifeguard system according to the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[0882] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0883] Step 1:

[0884] A user accesses the Internet using a device. This initiates online activities such as search queries, posting, and receiving / sending emails. The user's online behavior data (search queries, social media postings, email content) is obtained as input. This becomes the initial input for the system.

[0885] Step 2:

[0886] The server monitors the user's online behavior in real time. Here, we use the UserActionMonitor tool to capture the user's behavior data. As input, the user's online behavior data is monitored and sent to the server as behavior data. As output, this data is saved as behavior data capture.

[0887] Step 3:

[0888] The server inputs the captured behavioral data into a generative AI model. This AI model has learned from past crime data and uses an algorithm to assess risk levels. The captured behavioral data is passed to the generative AI model as input. The output is an assessment of the risk level for the behavioral data.

[0889] Step 4:

[0890] The server makes a decision based on the risk level assessed by the generative AI model. A warning message is generated for actions with a high risk level. Here, the WarningGenerator tool is used. The input is the risk level assessment result. The output is a generated warning message.

[0891] Step 5:

[0892] The server notifies the user's terminal of the generated warning message. Here, we use the NotificationManager tool. As input, there is the generated warning message, which is sent to the user's terminal. As output, the warning message is displayed on the user's terminal.

[0893] Step 6:

[0894] The server calculates and provides estimated damage amounts and information on where to seek advice for high-risk behavior. The inputs are behavioral data and risk level assessment results. The output is a calculated damage amount and information on where to seek advice, which are provided to the user.

[0895] Step 7:

[0896] It monitors and records how users react to warnings and provided information. The server collects this reaction data and uses it as training data for the generative model. As input, user reaction data is collected. As output, this data is saved as feedback for the generative model.

[0897] These steps enable the Digital Lifeguard system to monitor users' online behavior in real time and provide appropriate warnings and information.

[0898] 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.

[0899] This paper describes a digital lifeguard system incorporating an emotion engine, which is an interactive service for users to reduce the risk of becoming involved in online crimes and victimization. The system also recognizes the user's emotional state and provides warnings and advice based on that state.

[0900] System Configuration

[0901] The system mainly consists of the following components:

[0902] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral and emotional data.

[0903] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[0904] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[0905] 4. Emotion Engine: A component that recognizes and analyzes user emotions in real time.

[0906] Program processing and operation explanation

[0907] Users begin online activities

[0908] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[0909] Capturing behavioral and emotional data

[0910] The device captures the user's online behavior and input from the camera and microphone. The emotion engine analyzes the user's emotions using the captured video and audio data. This data is sent to the server.

[0911] Risk and Sentiment Analysis

[0912] The server receives and analyzes the behavioral and emotional data. The behavioral data is input into a generative model to assess the risk level. The emotional data is analyzed by an emotion engine to assess the user's current emotional state (e.g., stress, relief, anxiety).

[0913] Generating Emotion-Based Alerts

[0914] The server generates an appropriate warning message based on the risk level and the evaluation of the user's emotional state. For example, if the user is feeling stressed, the server can issue a warning in a gentle tone that takes into account the user's psychological state. A specific example of a message that could be generated is, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time."

[0915] Displaying warnings and providing countermeasure information

[0916] The server sends the generated warning message to the user's device. The device displays the warning message in real time, along with the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization for advice."

[0917] User Reactions and Feedback

[0918] The server records how users respond to the alerts and how their emotional state changes. These data are used as training data for the generative model and emotion engine, and are fed back to improve the accuracy of the entire system.

[0919] Specific examples

[0920] Example 1: If a user posts on social media, "I want to buy expensive branded goods at a discount," the server will scan the content and the user's emotions to detect the risk of fraud. Also, if the user's emotional state is "excited," the server will display a warning in a gentle tone to help them regain their composure.

[0921] Example 2: If a user receives an email containing a suspicious link, the server analyzes the email and, if it detects that the user's emotional state is "anxiety," it suggests ways to mitigate the risk and provides a message to help the user stay calm.

[0922] In this way, the digital lifeguard system of the present invention, which is combined with an emotion engine, more effectively provides an environment in which users can safely engage online, and realizes risk management that takes into account their emotional state.

[0923] The processing flow will be explained below.

[0924] Step 1:

[0925] The user begins their online journey. They use their device to access the Internet and search, post, and communicate. For example, they search for "cheapest brand name products."

[0926] Step 2:

[0927] The device captures the user's online behavior, such as search queries, social media posts, and received emails, and sends this data to a server.

[0928] Step 3:

[0929] The device's camera and microphone are used to capture emotional data from the user's facial expressions and voice, and this data is also sent to the server.

[0930] Step 4:

[0931] The server receives behavioral and emotional data sent from the device in real time, and performs privacy-conscious data filtering at this stage.

[0932] Step 5:

[0933] The server inputs the received behavioral data into a generative model, which is trained on historical crime data and fraud patterns and uses this data to assess risk levels. For example, a search query for "cheapest branded goods" might be deemed to have a high risk of fraud.

[0934] Step 6:

[0935] The server analyzes the emotion data using an emotion engine, evaluating the user's emotional state based on their facial expressions and voice, and detecting, for example, "anxiety" or "excitement."

[0936] Step 7:

[0937] The server combines the risk level assessment results with the emotional state assessment results to generate an appropriate warning message. If the risk is high and the user is feeling "anxious," it creates a gentler warning that takes into consideration the user's feelings. For example, it might say, "This search result may include fraudulent sites selling counterfeit branded goods. Please check with authorized retailers when you have time."

[0938] Step 8:

[0939] The server sends the generated warning message to the user's device. At the same time, it also generates an estimated loss amount and information on where to seek advice. For example, "If you fall victim to a scam, you can expect to lose 200,000 yen. Please contact the following organizations for advice."

[0940] Step 9:

[0941] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[0942] Step 10:

[0943] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[0944] Step 11:

[0945] The server records how the user responded to the warning, for example, whether the warning was ignored or acted upon, and also records the user's emotional state in response to the warning.

[0946] Step 12:

[0947] The server uses the recorded behavioral and emotional data as training data for the generative model and emotion engine, improving the accuracy of the entire system and adjusting it to ensure more accurate risk and emotion analysis from the next time onwards.

[0948] Through the above process, by combining an emotion engine with the digital lifeguard system, user safety can be ensured more effectively.

[0949] Example 2

[0950] 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."

[0951] In today's digital society, users are at a greater risk of becoming involved in crimes and fraud online. Furthermore, warnings and advice that ignore users' emotional state make it difficult to encourage appropriate responses. As a result, users' safety may not be adequately ensured, and the damage may worsen.

[0952] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing user behavioral data and emotional data and transmitting them to the server, means for inputting the user behavioral data into a generative model to evaluate the risk level, and means for analyzing the emotional data using an emotion engine to evaluate the user's emotional state. This ensures the safety of the user and enables appropriate warnings and advice that take into account the user's emotional state.

[0953] "Means for monitoring users' online behavior" refers to the ability to track and record users' activities on the Internet, such as searches, posts, and clicks, in real time.

[0954] "Means for capturing user behavioral data and emotional data and transmitting it to the server" refers to a function that collects the user's operation history, facial expressions, tone of voice, etc., and transfers this data to the server.

[0955] The "means of inputting into a generative model to assess risk levels" refers to a function that inputs acquired behavioral data into an AI model and performs risk assessment based on past crime data.

[0956] "Means for analyzing emotional data using an emotion engine and assessing the user's emotional state" is a function that analyzes captured video and audio data and determines the user's emotional state.

[0957] "Means for generating an appropriate warning message based on the risk level and emotional state and sending it to the user's device" refers to a function that creates an appropriate warning message for the user based on the risk assessment and emotional state, and immediately sends it to the user's device.

[0958] "Means for providing users with estimated damage amounts and information on consultation centers" is a function that provides users with information on appropriate consultation centers along with estimated damage amounts when a risk such as fraud is detected.

[0959] "Means for recording user responses and updating the AI ​​model" refers to a function that records user responses to warning messages and uses that data to improve the accuracy and efficiency of the AI ​​model.

[0960] A "generative model" is an artificial intelligence model that learns from past crime data and compares it with current data to perform risk assessment.

[0961] An "emotion engine" is a system element that analyzes emotional data such as images and sounds and determines the user's emotional state.

[0962] The digital lifeguard system of this invention ensures the safety of users online and provides appropriate warnings and advice taking into account their emotional state. This system mainly consists of four components: a server, a terminal, a generative model, and an emotion engine.

[0963] The server is the central part of the system and plays a key role in analyzing users' behavioral and emotional data. The terminal is a device (such as a PC, smartphone, or tablet) that users use to access the Internet, and is responsible for capturing behavioral and emotional data and sending it to the server. The generative model is installed in the server and is an AI model used to perform risk analysis based on past crime data. The emotion engine is a component that recognizes and analyzes users' emotions in real time.

[0964] The system begins operation when a user accesses the Internet using a device and begins searching, posting, or communicating. The device captures the user's online behavior and simultaneously captures input data from a camera and microphone. For example, the device's camera captures the user's facial expressions and the microphone records the user's tone of voice. This data is analyzed by an emotion engine to evaluate the user's emotional state.

[0965] The server receives the behavioral and emotional data sent from the device and uses it to analyze each. The behavioral data is input into a generative model to evaluate the risk level. For example, if a user attempts to click on a suspicious link, it is compared with a past database to determine whether the risk is high. The emotional data is analyzed by an emotion engine to evaluate the user's current emotional state. For example, it analyzes whether the user is feeling stressed while searching.

[0966] The server generates an appropriate warning message based on the risk level and the user's emotional state. If the user is feeling stressed, the server can issue a gentle warning that takes into account the user's psychological state. For example, a message such as "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time" can be generated.

[0967] The server sends the generated warning message to the user's device, which displays it in real time. It also displays the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization."

[0968] The server records how the user responded to the warning and also records how their emotional state changed. This data is used as training data for the generative model and emotion engine, and is fed back to improve the accuracy of the entire system.

[0969] Examples of prompt statements

[0970] Examples of prompts include:

[0971] 1. "Generate warning messages to reduce the risk of users being scammed on social media. Include specific messages for when users are excited."

[0972] 2. "When a user receives a suspicious email, please generate a message that provides appropriate advice in case they feel uneasy."

[0973] The above is an embodiment of the digital lifeguard system according to the present invention. This system provides an environment in which users can safely engage in online activities and realizes risk management that takes into account their emotional state.

[0974] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0975] Step 1: User begins online activity

[0976] A user accesses the Internet using a device and begins searching, posting, and communicating. Specifically, the user opens a browser, types "cheapest brand name products" in the search bar, and clicks the search button. This causes the device to begin recording the user's actions (input).

[0977] Step 2: Capture behavioral and emotional data

[0978] The device captures the user's behavioral data (browser history, click patterns, keyboard input, etc.). At the same time, the device's camera and microphone are used to capture the user's video and audio data (input). Specifically, the device's camera records the user's facial expressions in real time, and the microphone collects the tone of voice. This data is then sent to the emotion engine (output).

[0979] Step 3: Sending behavioral and emotional data

[0980] The device encrypts the captured behavioral and emotional data (data processing) and transmits it to the server (output), which then receives the data for analysis.

[0981] Step 4: Risk analysis

[0982] The server inputs the received behavioral data into the generative model and evaluates the risk level (data calculation). For example, if a user attempts to click on a suspicious link, the generative model compares it with a database of past crimes and determines that it is a high risk. It then generates a risk level (output).

[0983] Step 5: Sentiment analysis

[0984] The server inputs the received emotion data into the emotion engine to evaluate the user's emotional state (data calculation). Specifically, the emotion engine analyzes facial expressions from the video data and evaluates the tone of voice from the audio data. This allows it to determine, for example, that the user is in a state of stress, and generates an emotional state (output).

[0985] Step 6: Create a warning based on risk and sentiment

[0986] The server generates an appropriate warning message based on the analyzed risk level and emotional state (data calculation). For example, if the user is feeling stressed, it creates a gentle warning message about the risk of fraud. It generates a message such as, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with authorized retailers when you have time." (output).

[0987] Step 7: Sending a warning message

[0988] The server sends the generated warning message to the user's terminal (output), which is then ready to display this warning in real time.

[0989] Step 8: Displaying warning messages and providing countermeasure information

[0990] The device displays the received warning message in real time and provides additional countermeasure information. For example, it may display on the screen, "If you fall victim to a scam, you are expected to lose 200,000 yen. Please contact the following organization for assistance," along with the estimated loss amount and information on where to seek assistance (output).

[0991] Step 9: Gather user reactions and feedback

[0992] The server records how users respond to warning messages and collects user reaction data (input). For example, it records the number of times users clicked links after receiving a warning, or the time it took them to close the page. This data is used as training data for the generative model and emotion engine, and is used to improve the system in future releases (data calculation and output).

[0993] (Application example 2)

[0994] 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."

[0995] In today's digital society, users are increasingly exposed to the risk of online crime and fraud. As a result, many people suffer financial losses and psychological stress. However, conventional security systems simply detect risks and do not consider whether the warning is appropriate for the user's emotional state, resulting in inappropriate timing and formatting of warnings. Therefore, there is a need for warnings that take into account the user's emotional state and deliver warnings in an appropriate tone.

[0996] 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.

[0997] In this invention, the server includes means for monitoring users' online behavior, means for analyzing users' behavioral data and assessing risk levels, means for displaying warnings for risky behavior, means including an emotion engine for analyzing users' emotional states, means for adjusting warning content based on the emotional states, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the AI ​​model. This makes it possible to not only detect online risks but also provide appropriate warnings and advice that take into account the users' emotional states.

[0998] "Online behavior monitoring means" refers to a function that records and tracks the activities that users perform on the Internet (searches, browsing, sending and receiving emails, etc.).

[0999] "Means for analyzing behavioral data" refers to a function for analyzing data collected from users' online behavior and assessing risk levels and trends.

[1000] The "warning display means" is a function for displaying a warning to the user about risky behavior based on the analysis results.

[1001] The "emotion engine" is a function that analyzes the user's emotional state in real time and uses that data to diagnose the risks of behavior and adjust warning content.

[1002] The "means for adjusting the warning content" is a function for appropriately changing the content and tone of the warning message based on the user's emotional state analyzed by the emotion engine.

[1003] "Means for providing information on estimated damages and consultation services" is a function for providing users with the estimated monetary damages due to risky behavior and information on appropriate consultation services.

[1004] The "user response recording means" is a function that records the user's response to the warning and uses that data to update the system's AI model and improve its accuracy.

[1005] The digital lifeguard system of the present invention is a system that mainly includes functions of monitoring users' online behavior, analyzing behavioral and emotional data, assessing risks, displaying warnings, providing estimated damage amounts and information on where to seek help, and recording users' responses. This system is designed to help users stay safe in the online environment.

[1006] System configuration

[1007] The system consists of the following main components:

[1008] 1. Server: This is the main component that receives and analyzes users' online behavioral data and emotional data. It is equipped with an emotion engine and generative AI model.

[1009] 2. Device: A device (e.g., PC, smartphone, tablet, etc.) that allows a user to access the Internet. The device is equipped with a camera and microphone to capture behavioral and emotional data.

[1010] 3. Generative AI model: An AI model that learns from past crime data and performs risk assessment based on user behavior data.

[1011] 4. Emotion Engine: This component analyzes the user's emotional state in real time and uses that data to adjust the content of warnings.

[1012] Hardware and Software Used

[1013] Hardware:

[1014] Device camera: Uses OpenCV for facial recognition and emotion analysis.

[1015] Device microphone: Used to analyze emotional state through voice analysis using Librosa.

[1016] software:

[1017] OpenCV: A library for capturing and processing video data.

[1018] Librosa: A library for capturing and analyzing audio data.

[1019] Keras: A framework for emotion recognition using deep learning models.

[1020] Joblib: A library for loading pre-trained generative models for risk assessment.

[1021] Requests: A library for communicating with the server.

[1022] Data processing and calculation

[1023] The server receives the user's online behavioral data and emotion data sent from the device, and processes the data in the following steps:

[1024] Behavioral Data Analysis: Generative AI models are used to assess a user's overall online behavior and calculate their risk level.

[1025] Emotional Data Analysis: The emotion engine analyzes the video and audio data captured from the camera and microphone to determine the user's emotional state.

[1026] Warning message generation: The server generates appropriate warning messages based on risk level and emotional state, adjusting their tone and content.

[1027] Provision of information: Provide users with estimated damage amounts and information on where to seek advice, and suggest specific countermeasures.

[1028] Specific examples

[1029] Example 1: If a user tries to access a fraudulent website, the system will detect the risk and, if it determines that the user is stressed, will display a gentle warning saying, "This site poses a risk of fraud. Do not click on this link."

[1030] Example prompt sentence:

[1031] "We want to develop an application that uses artificial intelligence to analyze emotions and warn users of risky behavior. We want to combine data from a user's online activity, camera, and microphone to perform real-time emotion analysis and risk assessment. We call this system a 'security assistant application,' and it will help users avoid becoming victims of fraud and crime."

[1032] Thus, the present invention provides a system that can appropriately manage the risks of online activities while taking into consideration the emotional state of the user.

[1033] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1034] Step 1: "User begins online activity"

[1035] A user accesses the Internet using a device and performs online activities such as searching, browsing, and sending and receiving emails. The input is the user's various online activities, which are captured by the device in real time. The output is the behavioral data captured by the device.

[1036] Step 2: Capture behavioral and emotional data

[1037] The device captures the user's online behavioral data and simultaneously captures emotional data in real time using video and audio data obtained from a camera and microphone. The input is the user's facial expression and audio data, which the device collects and sends to the server. The output is the behavioral and emotional data sent to the server.

[1038] Step 3: "Behavioral Data Risk Assessment"

[1039] The server inputs the behavioral data sent from the device into a generative AI model and evaluates the risk level based on past crime data. Specifically, the generative AI model analyzes the behavioral data and determines whether there is a risk of fraud or crime. The input is the behavioral data sent to the server, and the output is the risk level assessment result.

[1040] Step 4: "Analyzing Emotional Data"

[1041] The server analyzes the emotion data using an emotion engine to evaluate the user's emotional state. The emotion engine performs facial recognition from video data and estimates emotions from audio data. The input is the video and audio data sent to the server, and the output is the user's emotional state (e.g., stress, anxiety, relief, etc.).

[1042] Step 5: Generate a warning message

[1043] The server generates an appropriate warning message based on the risk level assessment results and emotional state assessment results. Specifically, if the risk level is high, the content and tone of the message are adjusted according to the user's emotional state. For example, if the user is feeling stressed, the warning is issued in a gentle tone. The input is the risk level assessment results and emotional state assessment results, and the output is a warning message.

[1044] Step 6: "Displaying warning messages and providing countermeasure information"

[1045] The terminal displays the warning message sent from the server in real time, along with the estimated loss amount and information on where to seek advice. Specifically, the terminal displays the message "There is a risk of fraud. Do not click on this link." It also provides information such as "If you fall victim to a fraud, you could lose 200,000 yen. Please contact the following organization." The input is the warning message sent from the server, and the output is the warning and countermeasure information displayed on the terminal.

[1046] Step 7: User Response and Feedback

[1047] The server records how the user responded to the warning (e.g., did not click on the link, ignored the warning). It also tracks changes in the user's emotional state and feeds this data back as learning data for the AI ​​model. The input is the user's reaction data and data on changes in emotional state, and the output is an updated AI model. This improves the accuracy of the entire system.

[1048] 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.

[1049] 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.

[1050] 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.

[1051] [Fourth embodiment]

[1052] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1053] 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.

[1054] 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).

[1055] 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.

[1056] 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.

[1057] 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).

[1058] 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.

[1059] 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.

[1060] 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.

[1061] 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.

[1062] 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.

[1063] 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.

[1064] 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."

[1065] The following describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive service for reducing the risk of users becoming involved in crimes or becoming victims online, and specifically operates in the following manner.

[1066] System Configuration

[1067] The system mainly consists of the following components:

[1068] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral data.

[1069] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[1070] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[1071] Program processing and operation explanation

[1072] Users begin online activities

[1073] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[1074] Behavioral Data Capture

[1075] The server captures user behavior data in real time. To protect user privacy, it does not collect data that can identify personal information. User search queries, postings, received emails, etc. are monitored.

[1076] Risk Analysis

[1077] The captured behavioral data is input into a generative model stored on the server. The generative model learns from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if the search query "cheapest brand name products" is determined to be likely to be related to fraud, it will be classified as high risk.

[1078] Displaying warnings

[1079] If the risk is deemed high, the server will generate an appropriate warning message for the user, which will be displayed on the user's device. For example, the message may read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[1080] Providing estimated damage amounts and information on where to seek advice

[1081] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[1082] User Reactions and Feedback

[1083] The server records whether the user ignores the warning or whether they act on it, and the recorded data is used as training data for the generative model and fed back to improve the accuracy of the entire system.

[1084] Specific examples

[1085] Example 1: If a user posts on social media, "I want to buy expensive brand-name goods at a discount," the server will scan the content and detect risks. If applicable, it will display a warning and provide an estimated loss amount and information on where to seek advice.

[1086] Example 2: When a user receives an email containing a suspicious link, the server analyzes the email and displays a warning if it determines that the risk is high. It also notifies the user of the damage that could result from clicking the link and urges them to take action immediately.

[1087] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[1088] The processing flow will be explained below.

[1089] Step 1:

[1090] The user begins their online journey. They use their device to access the Internet and search, write, and communicate. For example, they enter a search query such as "cheapest brand name products."

[1091] Step 2:

[1092] The device captures the user's online behavior, including search queries, social media posts, and received emails, and sends this data to a server.

[1093] Step 3:

[1094] The server receives the behavioral data sent from the device in real time, where it undergoes pre-processing and privacy-conscious data filtering.

[1095] Step 4:

[1096] The server inputs the received data into an internal generative model, which has learned from past crime data and fraud patterns to assess the risk level. For example, it analyzes search results for "cheapest brand name products" and determines that there is a high probability of fraud.

[1097] Step 5:

[1098] The server receives the evaluation results from the generative model and determines the risk level. It is classified as high risk, medium risk, or low risk. If there is a high possibility of fraud, it is classified as high risk.

[1099] Step 6:

[1100] If the server determines that the risk is high, it will generate an appropriate warning message for the user, which will include specific risks and countermeasures.

[1101] Step 7:

[1102] The server generates a warning message and sends it to the user's device, such as "This search result may contain fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[1103] Step 8:

[1104] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[1105] Step 9:

[1106] The server generates a warning message along with estimated damages and information on where to seek advice, and provides it to the user. For example, it might say, "If you fall victim to a scam, you are likely to lose 200,000 yen. Please contact the following organization for advice."

[1107] Step 10:

[1108] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[1109] Step 11:

[1110] The server records how the user responded to the warning, for example, by storing a log of whether the warning was ignored, acted upon, or whether a helpline was contacted.

[1111] Step 12:

[1112] The server uses the recorded data as training data for the generative model, and the feedback improves the accuracy of the generative model, adjusting it to make future risk analyses more accurate.

[1113] Through the above process, the Digital Lifeguard System effectively reduces the risk of online crime and fraud, providing users with an environment where they can use the Internet with peace of mind.

[1114] Example 1

[1115] 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."

[1116] As the number of Internet users increases daily, the risk of becoming involved in online crimes and fraud is also increasing. In particular, many users suffer financial losses due to clicking on fraudulent links or visiting fraudulent websites. However, existing security measures have difficulty monitoring users' online behavior in real time and immediately warning them of risks, so effective countermeasures are needed.

[1117] 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.

[1118] In this invention, the server includes means for monitoring users' online behavior, means for capturing and preprocessing user behavior data in real time, means for inputting the captured data into a generative AI model and evaluating the risk level, means for displaying a warning for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the generative AI model. This provides an environment in which users can act safely online, and has the effect of significantly reducing the risk of crime and fraud.

[1119] "Online behavior" refers to a series of activities that users perform on the Internet, such as searching, posting, and receiving emails.

[1120] "Behavioral Data" is data that contains information about your online behavior, such as search queries, posts, and received emails.

[1121] "Real-time" means that data is processed as soon as it is generated, allowing for immediate analysis and evaluation without delay.

[1122] "Capture" is the act of collecting data about a user's online behavior.

[1123] "Preprocessing" is the process of removing unnecessary information from the captured raw data and converting it into a form suitable for analysis.

[1124] A "generative AI model" is an artificial intelligence algorithm that learns from past crime data and assesses risk based on new data.

[1125] A "risk level" is a number or category that evaluates the degree of danger associated with a user's behavior.

[1126] A "warning message" is a notification that alerts users to risky behavior.

[1127] "Estimated damage amount" is the amount of financial loss that a user is expected to incur if they fall victim to fraud or crime.

[1128] "Consultation information" is contact information for institutions or organizations that users should contact if they become victims.

[1129] "Recording a response" means recording how the user responded to the warning (whether they ignored it or followed it).

[1130] "Updating the AI ​​model" is the process of relearning the artificial intelligence algorithm based on recorded reaction data to improve its accuracy.

[1131] MODE FOR CARRYING OUT THE INVENTION

[1132] This invention is a system for reducing the risk of being involved in online crime and fraud by monitoring users' online behavior, assessing risks, and providing appropriate warning messages. The system mainly consists of three main components: a server, a user's device, and a generative AI model.

[1133] System Configuration

[1134] 1. Server:

[1135] The server captures and preprocesses users' online behavioral data in real time. It is preferable to use cloud services such as AWS (Amazon Web Services) or Google Cloud Platform. The server also contains a generative AI model built using machine learning libraries such as TensorFlow and PyTorch. This model evaluates risk levels based on the captured data.

[1136] 2. Terminal:

[1137] A terminal is a device that a user uses to access the internet, such as a PC, smartphone, or tablet. The terminal is responsible for displaying warning messages and consultation information sent from the server.

[1138] 3. Generative AI Model:

[1139] Generative AI models are trained using machine learning libraries such as TensorFlow and PyTorch based on historical crime data and fraud patterns, and then analyze captured user behavior data to assess risk levels.

[1140] Specific examples

[1141] Example 1: Risk assessment of search queries

[1142] When a user searches for "cheapest branded goods" on their device, the server captures the search query in real time. The captured data is input into a generative AI model, which assesses the risk based on past fraud patterns. If the risk is determined to be high, the server generates a warning message that reads, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers." and displays it on the device.

[1143] Example 2: Analysis of suspicious emails

[1144] When a user receives an email containing a suspicious link, the server captures the email in real time and performs text analysis. If the generative AI model determines that the email poses a high risk, the server generates a warning message that reads, "This email may be fraudulent. Do not click on the link." and displays it on the device.

[1145] Prompt Sentence Examples

[1146] "Analyze current user behavior based on past fraud data, conduct risk assessments, and generate warning messages if deemed high risk. Also consider countermeasures if users ignore the messages."

[1147] In this way, the digital lifeguard system of the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[1148] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1149] Step 1:

[1150] A user accesses the Internet. Using a device (PC, smartphone, tablet, etc.), the user opens a web browser and searches for, for example, "cheapest brand-name products."

[1151] Input: search query "cheapest brand name products".

[1152] Output: The user's search query is retrieved.

[1153] Step 2:

[1154] The server captures user behavior data in real time, including the search queries users type, the content they post, and the emails they receive.

[1155] Input: Your online behavior data (search queries, posts, received emails, etc.).

[1156] Output: Captured behavioral data.

[1157] Step 3:

[1158] The server preprocesses the captured behavioral data. This involves filtering unnecessary information from the text data and converting it into a format suitable for analysis. For example, natural language processing (NLP) techniques are used to extract important keywords.

[1159] Input: Raw captured data.

[1160] Output: Preprocessed data (data from which important keywords have been extracted).

[1161] Step 4:

[1162] The server then inputs the preprocessed data into a generative AI model, which uses historical crime data and fraud patterns to assess risk levels. For example, it might determine that a query like "cheapest brand name products" is likely to be fraud-related.

[1163] Input: Preprocessed data.

[1164] Output: Risk level assessment result (high risk, low risk, etc.).

[1165] Step 5:

[1166] If the server determines that the risk is high, it generates a warning message, which is then displayed on the user's device. For example, the message might read, "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from authorized retailers."

[1167] Input: Risk level assessment result (high risk).

[1168] Output: A warning message.

[1169] Step 6:

[1170] The server calculates the estimated loss if you fall victim to fraud or crime and provides it to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[1171] Input: Risk level assessment result (high risk).

[1172] Output: Estimated damage amount and consultation information.

[1173] Step 7:

[1174] The server records whether the user ignored or acted on the warning. This record includes the user's actions after the warning (whether they clicked on a link, acted on the warning, etc.). The recorded data is used as training data for the generative AI model.

[1175] Input: User response data.

[1176] Output: An updated generative AI model.

[1177] In this way, the system monitors users' online behavior in real time and issues alerts when necessary, reducing the risk of crime and fraud.

[1178] (Application example 1)

[1179] 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."

[1180] In modern society, the number of internet-based frauds and crimes is on the rise, creating an urgent need to provide an environment where users can safely engage online. However, current security services lack real-time monitoring of online behavior and immediate warnings, limiting their effectiveness in preventing damage. Furthermore, they lack a feedback mechanism that allows the system to evolve based on user responses, leaving challenges in improving the overall accuracy of the system. To address these challenges, a comprehensive system with real-time monitoring, immediate warnings, and a feedback mechanism is needed.

[1181] 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.

[1182] In this invention, the server includes means for monitoring users' online behavior, means for analyzing user behavior data and assessing risk levels, means for displaying warnings for risky behavior, means for providing users with estimated damage amounts and information on where to seek advice, means for recording users' responses and updating the generative model, means for monitoring user behavior in real time, and means for analyzing risks in real time and displaying immediate warnings. This provides an environment where users can engage in online activities with peace of mind, making it possible to significantly reduce the risk of crime and fraud.

[1183] "User" means any person or entity that uses the System to conduct online activities.

[1184] "Online behavior" is a general term for actions such as searching, posting, and communication that are carried out using the Internet.

[1185] A "server" is a central component of the system and is an information processing device that collects, analyzes, and stores user behavior data.

[1186] "Behavioral Data" is a record of certain actions you take online, including search queries, postings, and the content of emails you receive and send.

[1187] "Risk level" is an indicator that shows the likelihood that a user's behavior will lead to crime or fraud.

[1188] A "warning" is a message that warns users against high-risk behavior.

[1189] "Estimated damage amount" is the amount of financial damage expected if a user falls victim to fraud or crime.

[1190] "Consultation information" refers to contact information for the police, consumer centers, etc. that users can consult if they are victims of a crime or fraud.

[1191] A "generative model" is an artificial intelligence model that learns from past crime data and is used to analyze the risks associated with user behavior.

[1192] A "feedback mechanism" is a mechanism for recording user responses and using that data to improve the accuracy of the generative model.

[1193] "Real-time monitoring" refers to the monitoring of users' online behavior in real time.

[1194] "Immediate warning" means that a warning message is displayed immediately when high-risk behavior is detected.

[1195] MODE FOR CARRYING OUT THE INVENTION

[1196] This paper describes an embodiment of a digital lifeguard system according to the present invention. This system is an interactive security service that reduces the risk of users becoming involved in online crimes or becoming victims. Below, each component of the system and its operation will be specifically described.

[1197] System Configuration

[1198] The system mainly consists of the following components:

[1199] 1. Server: This is the main component that monitors users' online behavior and analyzes the behavioral data.

[1200] 2. Terminal: A device (personal computer, smartphone, tablet, etc.) that a user uses to access the Internet.

[1201] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[1202] Program processing and operation explanation

[1203] Monitoring your online behavior

[1204] Users access the Internet using their devices to search, post, and communicate. For example, suppose a user searches for "cheapest brand-name products." All of this behavior is monitored in real time.

[1205] Behavioral data capture and analysis

[1206] The server captures user behavior data in real time and inputs this data into a generative model. The generative model has learned from past crime data and fraud patterns and evaluates the risk level of the user's behavior. For example, if a search query such as "cheapest brand name products" is determined to be highly likely to be related to fraud, it will be classified as high risk.

[1207] Risk analysis and warning display

[1208] If the risk is determined to be high in real time, the server generates an appropriate warning message for the user and displays it on the device, such as "This search result may include fraudulent sites selling counterfeit branded goods. We recommend purchasing from an authorized retailer."

[1209] Providing estimated damage amounts and information on where to seek advice

[1210] Furthermore, the server calculates the estimated loss if you fall victim to fraud or crime and provides this to the user. At the same time, it also displays information on where to go for help (contact information for the police, consumer centers, etc.). For example, it might say, "If you fall victim to fraud, you can expect to lose 200,000 yen. Please contact the following organizations."

[1211] User Reactions and Feedback

[1212] The server records whether the user ignores the warning or whether they act on it. The recorded data is used as training data for the generative model and as feedback to improve the accuracy of the entire system.

[1213] Hardware and software used

[1214] Hardware: Personal computers, smartphones, tablets

[1215] software:

[1216] AI Library: Used to build generative AI models and perform risk analysis

[1217] Monitoring Tools: Monitor your users' online behavior with UserActionMonitor

[1218] Notification Management System: Display alerts using NotificationManager

[1219] Examples and prompts

[1220] Specific examples

[1221] If a user posts on social media that they want to buy expensive branded goods at a discount, the server will scan the post and detect any risks. If applicable, it will display a warning, provide an estimated loss amount, and provide information on where to seek advice.

[1222] Prompt Sentence Examples

[1223] "When a user searches for 'cheapest branded goods,' please generate a prompt to check for fraud risk."

[1224] As described above, the digital lifeguard system according to the present invention provides an environment in which users can safely engage in online activities, and significantly reduces the risk of crime and fraud.

[1225] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1226] Step 1:

[1227] A user accesses the Internet using a device. This initiates online activities such as search queries, posting, and receiving / sending emails. The user's online behavior data (search queries, social media postings, email content) is obtained as input. This becomes the initial input for the system.

[1228] Step 2:

[1229] The server monitors the user's online behavior in real time. Here, we use the UserActionMonitor tool to capture the user's behavior data. As input, the user's online behavior data is monitored and sent to the server as behavior data. As output, this data is saved as behavior data capture.

[1230] Step 3:

[1231] The server inputs the captured behavioral data into a generative AI model. This AI model has learned from past crime data and uses an algorithm to assess risk levels. The captured behavioral data is passed to the generative AI model as input. The output is an assessment of the risk level for the behavioral data.

[1232] Step 4:

[1233] The server makes a decision based on the risk level assessed by the generative AI model. A warning message is generated for actions with a high risk level. Here, the WarningGenerator tool is used. The input is the risk level assessment result. The output is a generated warning message.

[1234] Step 5:

[1235] The server notifies the user's terminal of the generated warning message. Here, we use the NotificationManager tool. As input, there is the generated warning message, which is sent to the user's terminal. As output, the warning message is displayed on the user's terminal.

[1236] Step 6:

[1237] The server calculates and provides estimated damage amounts and information on where to seek advice for high-risk behavior. The inputs are behavioral data and risk level assessment results. The output is a calculated damage amount and information on where to seek advice, which are provided to the user.

[1238] Step 7:

[1239] It monitors and records how users react to warnings and provided information. The server collects this reaction data and uses it as training data for the generative model. As input, user reaction data is collected. As output, this data is saved as feedback for the generative model.

[1240] These steps enable the Digital Lifeguard system to monitor users' online behavior in real time and provide appropriate warnings and information.

[1241] 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.

[1242] This paper describes a digital lifeguard system incorporating an emotion engine, which is an interactive service for users to reduce the risk of becoming involved in online crimes and victimization. The system also recognizes the user's emotional state and provides warnings and advice based on that state.

[1243] System Configuration

[1244] The system mainly consists of the following components:

[1245] 1. Server: The main component responsible for monitoring users' online behavior and analyzing behavioral and emotional data.

[1246] 2. Terminal: A device (PC, smartphone, tablet, etc.) that a user uses to access the Internet.

[1247] 3. Generative model: An AI model installed on the server that performs risk analysis based on past crime data.

[1248] 4. Emotion Engine: A component that recognizes and analyzes user emotions in real time.

[1249] Program processing and operation explanation

[1250] Users begin online activities

[1251] Users use their devices to access the Internet and perform searches, posts, and communications. For example, suppose a user searches for "cheapest prices on brand-name products."

[1252] Capturing behavioral and emotional data

[1253] The device captures the user's online behavior and input from the camera and microphone. The emotion engine analyzes the user's emotions using the captured video and audio data. This data is sent to the server.

[1254] Risk and Sentiment Analysis

[1255] The server receives and analyzes the behavioral and emotional data. The behavioral data is input into a generative model to assess the risk level. The emotional data is analyzed by an emotion engine to assess the user's current emotional state (e.g., stress, relief, anxiety).

[1256] Generating Emotion-Based Alerts

[1257] The server generates an appropriate warning message based on the risk level and the evaluation of the user's emotional state. For example, if the user is feeling stressed, the server can issue a warning in a gentle tone that takes into account the user's psychological state. A specific example of a message that could be generated is, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time."

[1258] Displaying warnings and providing countermeasure information

[1259] The server sends the generated warning message to the user's device. The device displays the warning message in real time, along with the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization for advice."

[1260] User Reactions and Feedback

[1261] The server records how users respond to the alerts and how their emotional state changes. These data are used as training data for the generative model and emotion engine, and are fed back to improve the accuracy of the entire system.

[1262] Specific examples

[1263] Example 1: If a user posts on social media, "I want to buy expensive branded goods at a discount," the server will scan the content and the user's emotions to detect the risk of fraud. Also, if the user's emotional state is "excited," the server will display a warning in a gentle tone to help them regain their composure.

[1264] Example 2: If a user receives an email containing a suspicious link, the server analyzes the email and, if it detects that the user's emotional state is "anxiety," it suggests ways to mitigate the risk and provides a message to help the user stay calm.

[1265] In this way, the digital lifeguard system of the present invention, which is combined with an emotion engine, more effectively provides an environment in which users can safely engage online, and realizes risk management that takes into account their emotional state.

[1266] The processing flow will be explained below.

[1267] Step 1:

[1268] The user begins their online journey. They use their device to access the Internet and search, post, and communicate. For example, they search for "cheapest brand name products."

[1269] Step 2:

[1270] The device captures the user's online behavior, such as search queries, social media posts, and received emails, and sends this data to a server.

[1271] Step 3:

[1272] The device's camera and microphone are used to capture emotional data from the user's facial expressions and voice, and this data is also sent to the server.

[1273] Step 4:

[1274] The server receives behavioral and emotional data sent from the device in real time, and performs privacy-conscious data filtering at this stage.

[1275] Step 5:

[1276] The server inputs the received behavioral data into a generative model, which is trained on historical crime data and fraud patterns and uses this data to assess risk levels. For example, a search query for "cheapest branded goods" might be deemed to have a high risk of fraud.

[1277] Step 6:

[1278] The server analyzes the emotion data using an emotion engine, evaluating the user's emotional state based on their facial expressions and voice, and detecting, for example, "anxiety" or "excitement."

[1279] Step 7:

[1280] The server combines the risk level assessment results with the emotional state assessment results to generate an appropriate warning message. If the risk is high and the user is feeling "anxious," it creates a gentler warning that takes into consideration the user's feelings. For example, it might say, "This search result may include fraudulent sites selling counterfeit branded goods. Please check with authorized retailers when you have time."

[1281] Step 8:

[1282] The server sends the generated warning message to the user's device. At the same time, it also generates an estimated loss amount and information on where to seek advice. For example, "If you fall victim to a scam, you can expect to lose 200,000 yen. Please contact the following organizations for advice."

[1283] Step 9:

[1284] The terminal displays warning messages sent from the server to the user in real time, in the form of dialog boxes or banners.

[1285] Step 10:

[1286] The device also displays the estimated damage amount and information on where to contact for help to the user, allowing the user to take appropriate action quickly.

[1287] Step 11:

[1288] The server records how the user responded to the warning, for example, whether the warning was ignored or acted upon, and also records the user's emotional state in response to the warning.

[1289] Step 12:

[1290] The server uses the recorded behavioral and emotional data as training data for the generative model and emotion engine, improving the accuracy of the entire system and adjusting it to ensure more accurate risk and emotion analysis from the next time onwards.

[1291] Through the above process, by combining an emotion engine with the digital lifeguard system, user safety can be ensured more effectively.

[1292] Example 2

[1293] 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."

[1294] In today's digital society, users are at a greater risk of becoming involved in crimes and fraud online. Furthermore, warnings and advice that ignore users' emotional state make it difficult to encourage appropriate responses. As a result, users' safety may not be adequately ensured, and the damage may worsen.

[1295] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing user behavioral data and emotional data and transmitting them to the server, means for inputting the user behavioral data into a generative model to evaluate the risk level, and means for analyzing the emotional data using an emotion engine to evaluate the user's emotional state. This ensures the safety of the user and enables appropriate warnings and advice that take into account the user's emotional state.

[1296] "Means for monitoring users' online behavior" refers to the ability to track and record users' activities on the Internet, such as searches, posts, and clicks, in real time.

[1297] "Means for capturing user behavioral data and emotional data and transmitting it to the server" refers to a function that collects the user's operation history, facial expressions, tone of voice, etc., and transfers this data to the server.

[1298] The "means of inputting into a generative model to assess risk levels" refers to a function that inputs acquired behavioral data into an AI model and performs risk assessment based on past crime data.

[1299] "Means for analyzing emotional data using an emotion engine and assessing the user's emotional state" is a function that analyzes captured video and audio data and determines the user's emotional state.

[1300] "Means for generating an appropriate warning message based on the risk level and emotional state and sending it to the user's device" refers to a function that creates an appropriate warning message for the user based on the risk assessment and emotional state, and immediately sends it to the user's device.

[1301] "Means for providing users with estimated damage amounts and information on consultation centers" is a function that provides users with information on appropriate consultation centers along with estimated damage amounts when a risk such as fraud is detected.

[1302] "Means for recording user responses and updating the AI ​​model" refers to a function that records user responses to warning messages and uses that data to improve the accuracy and efficiency of the AI ​​model.

[1303] A "generative model" is an artificial intelligence model that learns from past crime data and compares it with current data to perform risk assessment.

[1304] An "emotion engine" is a system element that analyzes emotional data such as images and sounds and determines the user's emotional state.

[1305] The digital lifeguard system of this invention ensures the safety of users online and provides appropriate warnings and advice taking into account their emotional state. This system mainly consists of four components: a server, a terminal, a generative model, and an emotion engine.

[1306] The server is the central part of the system and plays a key role in analyzing users' behavioral and emotional data. The terminal is a device (such as a PC, smartphone, or tablet) that users use to access the Internet, and is responsible for capturing behavioral and emotional data and sending it to the server. The generative model is installed in the server and is an AI model used to perform risk analysis based on past crime data. The emotion engine is a component that recognizes and analyzes users' emotions in real time.

[1307] The system begins operation when a user accesses the Internet using a device and begins searching, posting, or communicating. The device captures the user's online behavior and simultaneously captures input data from a camera and microphone. For example, the device's camera captures the user's facial expressions and the microphone records the user's tone of voice. This data is analyzed by an emotion engine to evaluate the user's emotional state.

[1308] The server receives the behavioral and emotional data sent from the device and uses it to analyze each. The behavioral data is input into a generative model to evaluate the risk level. For example, if a user attempts to click on a suspicious link, it is compared with a past database to determine whether the risk is high. The emotional data is analyzed by an emotion engine to evaluate the user's current emotional state. For example, it analyzes whether the user is feeling stressed while searching.

[1309] The server generates an appropriate warning message based on the risk level and the user's emotional state. If the user is feeling stressed, the server can issue a gentle warning that takes into account the user's psychological state. For example, a message such as "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with an authorized retailer when you have time" can be generated.

[1310] The server sends the generated warning message to the user's device, which displays it in real time. It also displays the estimated loss amount and information on where to seek advice. For example, it might say, "If you fall victim to a scam, you could lose 200,000 yen. Please contact the following organization."

[1311] The server records how the user responded to the warning and also records how their emotional state changed. This data is used as training data for the generative model and emotion engine, and is fed back to improve the accuracy of the entire system.

[1312] Examples of prompt statements

[1313] Examples of prompts include:

[1314] 1. "Generate warning messages to reduce the risk of users being scammed on social media. Include specific messages for when users are excited."

[1315] 2. "When a user receives a suspicious email, please generate a message that provides appropriate advice in case they feel uneasy."

[1316] The above is an embodiment of the digital lifeguard system according to the present invention. This system provides an environment in which users can safely engage in online activities and realizes risk management that takes into account their emotional state.

[1317] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1318] Step 1: User begins online activity

[1319] A user accesses the Internet using a device and begins searching, posting, and communicating. Specifically, the user opens a browser, types "cheapest brand name products" in the search bar, and clicks the search button. This causes the device to begin recording the user's actions (input).

[1320] Step 2: Capture behavioral and emotional data

[1321] The device captures the user's behavioral data (browser history, click patterns, keyboard input, etc.). At the same time, the device's camera and microphone are used to capture the user's video and audio data (input). Specifically, the device's camera records the user's facial expressions in real time, and the microphone collects the tone of voice. This data is then sent to the emotion engine (output).

[1322] Step 3: Sending behavioral and emotional data

[1323] The device encrypts the captured behavioral and emotional data (data processing) and transmits it to the server (output), which then receives the data for analysis.

[1324] Step 4: Risk analysis

[1325] The server inputs the received behavioral data into the generative model and evaluates the risk level (data calculation). For example, if a user attempts to click on a suspicious link, the generative model compares it with a database of past crimes and determines that it is a high risk. It then generates a risk level (output).

[1326] Step 5: Sentiment analysis

[1327] The server inputs the received emotion data into the emotion engine to evaluate the user's emotional state (data calculation). Specifically, the emotion engine analyzes facial expressions from the video data and evaluates the tone of voice from the audio data. This allows it to determine, for example, that the user is in a state of stress, and generates an emotional state (output).

[1328] Step 6: Create a warning based on risk and sentiment

[1329] The server generates an appropriate warning message based on the analyzed risk level and emotional state (data calculation). For example, if the user is feeling stressed, it creates a gentle warning message about the risk of fraud. It generates a message such as, "This search result may include fraudulent sites selling counterfeit brand-name products. Please check with authorized retailers when you have time." (output).

[1330] Step 7: Sending a warning message

[1331] The server sends the generated warning message to the user's terminal (output), which is then ready to display this warning in real time.

[1332] Step 8: Displaying warning messages and providing countermeasure information

[1333] The device displays the received warning message in real time and provides additional countermeasure information. For example, it may display on the screen, "If you fall victim to a scam, you are expected to lose 200,000 yen. Please contact the following organization for assistance," along with the estimated loss amount and information on where to seek assistance (output).

[1334] Step 9: Gather user reactions and feedback

[1335] The server records how users respond to warning messages and collects user reaction data (input). For example, it records the number of times users clicked links after receiving a warning, or the time it took them to close the page. This data is used as training data for the generative model and emotion engine, and is used to improve the system in future releases (data calculation and output).

[1336] (Application example 2)

[1337] 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."

[1338] In today's digital society, users are increasingly exposed to the risk of online crime and fraud. As a result, many people suffer financial losses and psychological stress. However, conventional security systems simply detect risks and do not consider whether the warning is appropriate for the user's emotional state, resulting in inappropriate timing and formatting of warnings. Therefore, there is a need for warnings that take into account the user's emotional state and deliver warnings in an appropriate tone.

[1339] 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.

[1340] In this invention, the server includes means for monitoring users' online behavior, means for analyzing users' behavioral data and assessing risk levels, means for displaying warnings for risky behavior, means including an emotion engine for analyzing users' emotional states, means for adjusting warning content based on the emotional states, means for providing users with estimated damage amounts and information on where to seek advice, and means for recording users' responses and updating the AI ​​model. This makes it possible to not only detect online risks but also provide appropriate warnings and advice that take into account the users' emotional states.

[1341] "Online behavior monitoring means" refers to a function that records and tracks the activities that users perform on the Internet (searches, browsing, sending and receiving emails, etc.).

[1342] "Means for analyzing behavioral data" refers to a function for analyzing data collected from users' online behavior and assessing risk levels and trends.

[1343] The "warning display means" is a function for displaying a warning to the user about risky behavior based on the analysis results.

[1344] The "emotion engine" is a function that analyzes the user's emotional state in real time and uses that data to diagnose the risks of behavior and adjust warning content.

[1345] The "means for adjusting the warning content" is a function for appropriately changing the content and tone of the warning message based on the user's emotional state analyzed by the emotion engine.

[1346] "Means for providing information on estimated damages and consultation services" is a function for providing users with the estimated monetary damages due to risky behavior and information on appropriate consultation services.

[1347] The "user response recording means" is a function that records the user's response to the warning and uses that data to update the system's AI model and improve its accuracy.

[1348] The digital lifeguard system of the present invention is a system that mainly includes functions of monitoring users' online behavior, analyzing behavioral and emotional data, assessing risks, displaying warnings, providing estimated damage amounts and information on where to seek help, and recording users' responses. This system is designed to help users stay safe in the online environment.

[1349] System configuration

[1350] The system consists of the following main components:

[1351] 1. Server: This is the main component that receives and analyzes users' online behavioral data and emotional data. It is equipped with an emotion engine and generative AI model.

[1352] 2. Device: A device (e.g., PC, smartphone, tablet, etc.) that allows a user to access the Internet. The device is equipped with a camera and microphone to capture behavioral and emotional data.

[1353] 3. Generative AI model: An AI model that learns from past crime data and performs risk assessment based on user behavior data.

[1354] 4. Emotion Engine: This component analyzes the user's emotional state in real time and uses that data to adjust the content of warnings.

[1355] Hardware and Software Used

[1356] Hardware:

[1357] Device camera: Uses OpenCV for facial recognition and emotion analysis.

[1358] Device microphone: Used to analyze emotional state through voice analysis using Librosa.

[1359] software:

[1360] OpenCV: A library for capturing and processing video data.

[1361] Librosa: A library for capturing and analyzing audio data.

[1362] Keras: A framework for emotion recognition using deep learning models.

[1363] Joblib: A library for loading pre-trained generative models for risk assessment.

[1364] Requests: A library for communicating with the server.

[1365] Data processing and calculation

[1366] The server receives the user's online behavioral data and emotion data sent from the device, and processes the data in the following steps:

[1367] Behavioral Data Analysis: Generative AI models are used to assess a user's overall online behavior and calculate their risk level.

[1368] Emotional Data Analysis: The emotion engine analyzes the video and audio data captured from the camera and microphone to determine the user's emotional state.

[1369] Warning message generation: The server generates appropriate warning messages based on risk level and emotional state, adjusting their tone and content.

[1370] Provision of information: Provide users with estimated damage amounts and information on where to seek advice, and suggest specific countermeasures.

[1371] Specific examples

[1372] Example 1: If a user tries to access a fraudulent website, the system will detect the risk and, if it determines that the user is stressed, will display a gentle warning saying, "This site poses a risk of fraud. Do not click on this link."

[1373] Example prompt sentence:

[1374] "We want to develop an application that uses artificial intelligence to analyze emotions and warn users of risky behavior. We want to combine data from a user's online activity, camera, and microphone to perform real-time emotion analysis and risk assessment. We call this system a 'security assistant application,' and it will help users avoid becoming victims of fraud and crime."

[1375] Thus, the present invention provides a system that can appropriately manage the risks of online activities while taking into consideration the emotional state of the user.

[1376] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1377] Step 1: "User begins online activity"

[1378] A user accesses the Internet using a device and performs online activities such as searching, browsing, and sending and receiving emails. The input is the user's various online activities, which are captured by the device in real time. The output is the behavioral data captured by the device.

[1379] Step 2: Capture behavioral and emotional data

[1380] The device captures the user's online behavioral data and simultaneously captures emotional data in real time using video and audio data obtained from a camera and microphone. The input is the user's facial expression and audio data, which the device collects and sends to the server. The output is the behavioral and emotional data sent to the server.

[1381] Step 3: "Behavioral Data Risk Assessment"

[1382] The server inputs the behavioral data sent from the device into a generative AI model and evaluates the risk level based on past crime data. Specifically, the generative AI model analyzes the behavioral data and determines whether there is a risk of fraud or crime. The input is the behavioral data sent to the server, and the output is the risk level assessment result.

[1383] Step 4: "Analyzing Emotional Data"

[1384] The server analyzes the emotion data using an emotion engine to evaluate the user's emotional state. The emotion engine performs facial recognition from video data and estimates emotions from audio data. The input is the video and audio data sent to the server, and the output is the user's emotional state (e.g., stress, anxiety, relief, etc.).

[1385] Step 5: Generate a warning message

[1386] The server generates an appropriate warning message based on the risk level assessment results and emotional state assessment results. Specifically, if the risk level is high, the content and tone of the message are adjusted according to the user's emotional state. For example, if the user is feeling stressed, the warning is issued in a gentle tone. The input is the risk level assessment results and emotional state assessment results, and the output is a warning message.

[1387] Step 6: "Displaying warning messages and providing countermeasure information"

[1388] The terminal displays the warning message sent from the server in real time, along with the estimated loss amount and information on where to seek advice. Specifically, the terminal displays the message "There is a risk of fraud. Do not click on this link." It also provides information such as "If you fall victim to a fraud, you could lose 200,000 yen. Please contact the following organization." The input is the warning message sent from the server, and the output is the warning and countermeasure information displayed on the terminal.

[1389] Step 7: User Response and Feedback

[1390] The server records how the user responded to the warning (e.g., did not click on the link, ignored the warning). It also tracks changes in the user's emotional state and feeds this data back as learning data for the AI ​​model. The input is the user's reaction data and data on changes in emotional state, and the output is an updated AI model. This improves the accuracy of the entire system.

[1391] 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.

[1392] 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.

[1393] 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.

[1394] 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.

[1395] 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.

[1396] 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.

[1397] 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).

[1398] 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.

[1399] 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."

[1400] 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.

[1401] 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).

[1402] 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.

[1403] 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.

[1404] 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.

[1405] 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.

[1406] 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.

[1407] 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.

[1408] 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.

[1409] 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.

[1410] 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.

[1411] 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.

[1412] The following is further disclosed regarding the above embodiment.

[1413] (Claim 1)

[1414] means for monitoring users' online activities;

[1415] A means for analyzing user behavior data and assessing risk levels;

[1416] a means of displaying warnings against risky behavior;

[1417] A means of providing users with estimated damage amounts and information on where to seek advice;

[1418] A means to record user responses and update the AI ​​model;

[1419] A system including:

[1420] (Claim 2)

[1421] The system of claim 1, which uses a generative model to learn from past crime data and perform risk analysis.

[1422] (Claim 3)

[1423] 2. The system according to claim 1, which analyzes user behavior data in real time and displays an immediate warning.

[1424] "Example 1"

[1425] (Claim 1)

[1426] means for monitoring users' online activities;

[1427] A means for capturing and pre-processing user behavior data in real time;

[1428] a means for inputting the captured data into a generative AI model to assess risk levels;

[1429] a means of displaying warnings against risky behavior;

[1430] A means of providing users with estimated damage amounts and information on where to seek advice;

[1431] A means of recording user responses and updating the generative AI model;

[1432] A system including:

[1433] (Claim 2)

[1434] The system of claim 1, which uses a generative AI model to learn from past crime data and perform risk analysis.

[1435] (Claim 3)

[1436] The system of claim 1, which analyzes user behavior data in real time and displays an immediate warning if there is a high risk.

[1437] "Application Example 1"

[1438] (Claim 1)

[1439] means for monitoring users' online activities;

[1440] A means for analyzing user behavior data and assessing risk levels;

[1441] a means of displaying warnings against risky behavior;

[1442] A means of providing users with estimated damage amounts and information on where to seek advice;

[1443] a means for recording user responses and updating the generative model;

[1444] a means for monitoring user behavior in real time;

[1445] A means to analyze risks in real time and display immediate warnings,

[1446] A system including:

[1447] (Claim 2)

[1448] The system of claim 1, which uses a generative model to learn from past crime data and perform risk analysis.

[1449] (Claim 3)

[1450] 2. The system according to claim 1, which analyzes user behavior data in real time and displays an immediate warning.

[1451] "Example 2: Combining Emotion Engines"

[1452] (Claim 1)

[1453] means for monitoring users' online activities;

[1454] means for capturing and transmitting user behavioral and emotional data to a server;

[1455] A means for inputting user behavior data into a generative model to assess risk levels;

[1456] means for analyzing the emotion data using an emotion engine to assess the user's emotional state;

[1457] means for generating an appropriate warning message based on the risk level and emotional state and sending the message to the user's device;

[1458] A means of providing users with estimated damage amounts and information on where to seek advice;

[1459] A means to record user responses and update the AI ​​model;

[1460] A system including:

[1461] (Claim 2)

[1462] The system of claim 1, which uses a generative model to learn from past crime data and perform risk analysis.

[1463] (Claim 3)

[1464] 2. The system according to claim 1, which analyzes user behavioral data and emotional data in real time and displays an immediate warning.

[1465] "Application example 2 when combining emotion engines"

[1466] (Claim 1)

[1467] means for monitoring users' online activities;

[1468] A means for analyzing user behavior data and assessing risk levels;

[1469] a means of displaying warnings against risky behavior;

[1470] means including an emotion engine for analyzing an emotional state of a user;

[1471] means for tailoring alert content based on emotional state;

[1472] A means of providing users with estimated damage amounts and information on where to seek advice;

[1473] A means to record user responses and update the AI ​​model;

[1474] A system including:

[1475] (Claim 2)

[1476] The system of claim 1, which uses a generative model to learn from past crime data and perform risk analysis.

[1477] (Claim 3)

[1478] 2. The system according to claim 1, which analyzes user behavior data in real time and displays an immediate warning. [Explanation of symbols]

[1479] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for monitoring users' online activities; A means for analyzing user behavior data and assessing risk levels; a means of displaying warnings against risky behavior; A means of providing users with estimated damage amounts and information on where to seek advice; A means to record user responses and update the AI ​​model; A system including:

2. The system according to claim 1, wherein the system uses a generative model to learn from past crime data and perform risk analysis.

3. 2. The system according to claim 1, wherein the system analyzes user behavior data in real time and immediately displays a warning.

Images