system

Abstract

We provide the system. [Solution] A system including a generative AI means for analyzing suspicious phone calls and emails received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious links and information provided by elderly people, and a means for issuing warnings and advice regarding the detected suspicious links and information.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method executed by at least one processor, the method including a step of receiving a user utterance, a step of adding the user utterance to a prompt including an instruction sentence related to a description of a character of the chatbot, a step of encoding the prompt, and a step of inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The risk of elderly people falling victim to fraud is increasing. In particular, fraud via suspicious telephone calls and emails is on the rise, and it is difficult for elderly people to properly judge and deal with these incidents.

Means for Solving the Problems

[0005] Generative AI is used to analyze suspicious telephone calls and emails received by elderly people, and to determine the possibility of fraud. Further, a system is provided that detects suspicious links and information provided by elderly people, and issues warnings and advice. This enables elderly people to protect themselves from fraud.

Brief Description of the Drawings

[0006] [Figure 1]This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 1 of Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2 of Embodiment 2. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Form Example 2. [Figure 15] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 3 of Example 3. [Figure 16] This is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Form Example 3. [Figure 17] This is a sequence diagram showing the processing flow of the data processing system in Example 1 of Embodiment 1 when an emotion engine is combined. [Figure 18] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Embodiment 1 when an emotion engine is combined. [Figure 19] This is a sequence diagram showing the processing flow of the data processing system in Example 2 of Embodiment 2 when an emotion engine is combined. [Figure 20] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Embodiment 2 when an emotion engine is combined. [Figure 21] This is a sequence diagram showing the processing flow of the data processing system in Example 3 of Embodiment 3 when an emotion engine is combined. [Figure 22] This is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Embodiment 3 when an emotion engine is combined. [Figure 23] This is a sequence diagram showing the processing flow of a data processing system in another embodiment. [Modes for Carrying Out the Invention]

[0007] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, terms used in the following description will be explained.

[0009] In the following embodiments, a referenced processor (hereinafter simply referred to as "processor") may be a single arithmetic device, or a combination of multiple arithmetic devices. Further, the processor may be a single type of arithmetic device, or a combination of multiple types of arithmetic devices. Examples of the arithmetic device include a CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), or TPU (TENSOR PROCESSING UNIT (registered trademark)), and the like.

[0010] In the following embodiments, a referenced RAM (Random Access Memory) is a memory in which information is temporarily stored, and is used as a work memory by a processor.

[0011] In the following embodiments, a referenced storage is one or more non-volatile storage devices that store various programs, various parameters, and the like. Examples of the non-volatile storage device include a flash memory (SSD (Solid State Drive)), a magnetic disk (for example, a hard disk), or a magnetic tape, and the like.

[0012] In the following embodiments, a referenced communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F governs communication between a plurality of 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), or Bluetooth (registered trademark), and the like.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0014] [First Embodiment]

[0015] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0016] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0017] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0018] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0019] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0022] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0026] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0027] "Example of form 1"

[0028] One embodiment of the present invention involves a generative AI system that analyzes suspicious phone calls and emails received by elderly individuals. This generative AI uses natural language processing technology to analyze the content of phone calls and emails and determine the likelihood of them being scams. Specifically, it analyzes keywords, phrases, and context contained in the content of phone calls and emails, and if it determines that there is a high probability of fraud, it warns the elderly individual with that information.

[0029] "Example of form 2"

[0030] Furthermore, there are methods to detect suspicious links and information provided by elderly individuals. These methods monitor the actions of elderly individuals when they use the internet, and if they attempt to provide suspicious links or information, they detect the action and warn the elderly individual. Specifically, they detect actions such as elderly individuals attempting to click on links in emails or access suspicious websites, and warn the elderly individual.

[0031] "Example of form 3"

[0032] Furthermore, there are mechanisms to provide warnings and advice regarding detected suspicious links and information. These mechanisms provide visual or auditory notifications to older adults and offer specific advice to protect themselves from fraud. Specifically, they may instruct older adults not to click on suspicious links or advise them to avoid accessing suspicious websites. These notifications and advice are provided in a way that is easy for older adults to understand and help them take specific actions to protect themselves from fraud.

[0033] The following describes the processing flow for each example of the form.

[0034] "Example of form 1"

[0035] Step 1: Input the content of suspicious phone calls or emails received by the elderly person into a generative AI.

[0036] Step 2: The generative AI uses natural language processing techniques to analyze the content of phone calls and emails. This analysis includes the analysis of keywords, phrases, and context.

[0037] Step 3: The generative AI determines the likelihood of fraud based on the analysis results. If it determines that there is a high probability of fraud, it warns the elderly about the information.

[0038] "Example of form 2"

[0039] Step 1: Monitor the actions taken by elderly individuals when they use the internet.

[0040] Step 2: Detect any attempts to provide suspicious links or information.

[0041] Step 3: Issue a warning to the elderly person regarding the detected operation.

[0042] "Example of form 3"

[0043] Step 1: If suspicious links or information are detected, provide visual or auditory notifications to the elderly.

[0044] Step 2: The notification will include specific advice on how to protect yourself from fraud.

[0045] Step 3: Provide notifications and advice in a way that is easy for seniors to understand, and help them take concrete actions to protect themselves from fraud.

[0046] (Example 1)

[0047] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0048] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to provide appropriate warnings. However, conventional methods have insufficient information processing capabilities for determining the potential for fraud, resulting in a high risk of elderly people becoming victims of fraud.

[0049] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0050] In this invention, the server includes information processing means, decision means, and identification means. This makes it possible to analyze suspicious communications received by the elderly, assess the possibility of fraud, and provide appropriate warnings.

[0051] "Information processing means" refers to a device or system that has the function of analyzing suspicious communications received by elderly people and extracting important keywords using natural language processing technology.

[0052] A "decision-making tool" is a device or system that has the function of evaluating the possibility of fraud based on the analyzed communication content and generating a warning message based on the result.

[0053] "Identification means" refers to a device or system that has the function of detecting suspicious information provided by an elderly person and identifying whether that information may be related to fraud.

[0054] A "notification means" is a device or system that has the function of issuing a warning about detected suspicious information and providing warning messages to elderly people in a visual or auditory form.

[0055] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and process human language, and is used to extract meaning from text data.

[0056] A "warning message" is a message generated to alert older adults to potentially fraudulent communications, and is provided visually or audibly.

[0057] This invention is a system for analyzing suspicious communications received by elderly people and evaluating the possibility of fraud. The system includes information processing means, judgment means, identification means, and notification means.

[0058] 1. Program generation

[0059] The user generates a generative AI program using natural language processing technology. This program has the function of analyzing suspicious communications received by elderly people and determining the possibility of fraud.

[0060] 2. Program Processing

[0061] The server uses Python's natural language processing libraries, NLTK and spaCy, to analyze the communication content. The server normalizes the text data and extracts important keywords and phrases. Furthermore, it uses a generative AI model to analyze the context and assess the likelihood of fraud.

[0062] 3. Specific examples

[0063] For example, if an elderly person receives an email saying, "Your bank account has been used fraudulently. Please contact us immediately," the server will detect keywords such as "bank account," "fraudulent use," and "urgent," and determine that it is highly likely to be a scam. Based on this information, the server will generate a warning message and notify the user via their device that "This email may be a scam. Please be careful."

[0064] 4. Example of a prompt statement

[0065] An example of a prompt message for a generative AI model is, "Please analyze the content of this email and assess its likelihood of being a scam."

[0066] In this way, the system can respond quickly and accurately to suspicious communications received by the elderly, thereby preventing them from becoming victims of fraud.

[0067] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0068] Step 1:

[0069] The user inputs the content of suspicious communications received by the elderly person into the terminal. The terminal then sends this input data to the server. The input data is in text format.

[0070] Step 2:

[0071] The server preprocesses the received text data. Specifically, it uses Python's natural language processing libraries, NLTK and spaCy, to normalize the text. This includes removing unnecessary symbols and spaces, and converting words to lowercase. As a result of the preprocessing, text data in a format suitable for analysis is output.

[0072] Step 3:

[0073] The server extracts important keywords and phrases from pre-processed text data. This includes using natural language processing techniques to identify words that may be related to fraud. For example, keywords such as "bank account," "fraudulent use," and "urgent" are extracted. The extracted keywords become input data for evaluating the likelihood of fraud.

[0074] Step 4:

[0075] The server uses a generative AI model to analyze the context based on extracted keywords. The AI ​​model understands the meaning of the entire text and assesses the likelihood of fraud. This assessment determines whether the text is likely to be fraudulent. The evaluation results become input data for generating a warning message.

[0076] Step 5:

[0077] The server generates a warning message based on the evaluation results. If it is determined that there is a high probability of fraud, a warning message is created. For example, a message such as "This email may be a scam. Please be careful." is generated. The generated message is sent to the terminal.

[0078] Step 6:

[0079] The device displays warning messages received from the server to the user. Users can receive warnings visually or audibly. This helps older adults to be alerted to potentially fraudulent communications.

[0080] (Application Example 1)

[0081] Next, we will describe Application Example 1 of Form 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."

[0082] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to issue appropriate warnings. However, conventional methods can lead to delays or incorrect judgments, putting the elderly at risk of becoming victims. To solve this problem, technology is needed that can analyze communication content in real time and determine the potential for fraud with high accuracy.

[0083] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0084] In this invention, the server includes information processing means, evaluation means, and identification means. This makes it possible to analyze received communications in real time and quickly display a warning if there is a possibility of fraud.

[0085] "Information processing means" refers to a device or program that analyzes the content of received communications and performs data processing to determine the possibility of fraud.

[0086] An "evaluation tool" is a device or program that scores the likelihood of fraud based on the analyzed communication content and makes a judgment based on the result.

[0087] "Identification means" refers to a device or program that detects suspicious elements from information provided by elderly individuals and issues a warning based on that detection.

[0088] "Notification means" refers to a device or program that provides a visual or auditory warning to a user in the event of potential fraud.

[0089] "Means of real-time analysis" refers to a device or program that instantly analyzes the content of a communication as soon as it is received and quickly outputs the results.

[0090] The system for implementing this invention involves a server and a terminal working in cooperation. The server is equipped with information processing means, evaluation means, and identification means, and analyzes the communication content received from the terminal. Specifically, the server uses a natural language processing library (e.g., spaCy, NLTK) to tokenize the communication content and extract keywords and phrases. Next, it uses a generative AI model (e.g., OpenAI® GPT) to analyze the context of the extracted keywords and phrases and score the likelihood of fraud.

[0091] The device receives analysis results from the server and, if it detects a high probability of fraud, provides the user with a visual or auditory warning. This allows the user to respond quickly to suspicious communications.

[0092] As a concrete example, if a user receives an email with the subject line "Important Notice from the Bank," the device sends its contents to the server. The server extracts keywords such as "bank," "important," and "notice," and analyzes them using a generative AI model. If it is determined that the email is highly likely to be fraudulent, the device displays a warning message saying, "This email may be fraudulent. Do not click on the link."

[0093] An example of a prompt message is: "Please analyze the following email content to determine if it is a scam. Email content: 'Important notice from your bank. Please update your account information.'"

[0094] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0095] Step 1:

[0096] The terminal retrieves the content of communications (phone calls and emails) received by the user. This content is sent to the server as text data. The input is the content of communications received by the user, and the output is the transmission of text data to the server.

[0097] Step 2:

[0098] The server tokenizes the received text data using a natural language processing library (e.g., spaCy, NLTK). This process extracts keywords and phrases from the text data. The input is the text data sent from the terminal, and the output is the extracted keywords and phrases.

[0099] Step 3:

[0100] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the context of the extracted keywords and phrases. This analysis scores the likelihood of fraud. The input is the extracted keywords and phrases, and the output is a score indicating the likelihood of fraud.

[0101] Step 4:

[0102] The server determines whether there is a high probability of fraud based on the scoring results. If it determines that there is a high probability of fraud, it generates a warning message. The input is the score indicating the likelihood of fraud, and the output is the warning message.

[0103] Step 5:

[0104] The terminal notifies the user of warning messages received from the server. These notifications are visual or auditory, warning the user of potential fraud. The input is the warning message from the server, and the output is the warning notification to the user.

[0105] (Example 2)

[0106] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0107] When elderly people use the internet, there is a risk of accessing suspicious links or information. Such risks can lead to phishing scams and malware infections, posing a threat to the safety of the elderly. Traditional methods are insufficient to prevent these risks, making it difficult for the elderly to use the internet with peace of mind.

[0108] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0109] In this invention, the server includes information processing means, monitoring means, and notification means. This makes it possible to immediately issue a warning when an elderly person attempts to access suspicious links or information, thereby preventing harm.

[0110] "Information processing means" refers to technology for analyzing suspicious communications received by elderly people and understanding their content.

[0111] "Monitoring measures" refer to technologies that monitor the actions of elderly people in real time and detect suspicious links or information.

[0112] "Notification means" refers to technology that generates warnings for detected suspicious links or information and informs elderly people.

[0113] A "monitoring device" is a device used to monitor the actions of elderly people in real time and has the function of recording the user's behavior.

[0114] "Analysis methods" refer to technologies used to compare suspicious links and information with a database and evaluate their safety.

[0115] "Display means" refers to technologies for displaying warning messages to the elderly visually or audibly.

[0116] This invention is a system for reducing the risk of elderly people accessing suspicious links or information when using the internet. The system includes information processing means, monitoring means, and notification means.

[0117] The server uses information processing tools to analyze suspicious communications received by elderly individuals. This analysis utilizes natural language processing technology to understand the content of the communications. Specifically, the server uses software such as a "natural language processing engine" to analyze the content of emails and messages.

[0118] The terminal uses monitoring devices to monitor the elderly person's actions in real time. Monitoring software installed on the terminal captures links the user attempts to click and URLs of websites they attempt to access, and sends them to a server. Security software is used for this monitoring.

[0119] The server uses analysis tools to compare URLs sent from the terminal against a database of known suspicious links and phishing sites. For example, it uses a "security evaluation API" to assess the security of a URL.

[0120] If suspicious links or information are detected, the server generates a warning message using notification mechanisms. This warning message is sent to the terminal and notified to the elderly visually or audibly through display mechanisms. For example, a message such as "This link is suspicious. Please stop accessing it." might be displayed.

[0121] As a concrete example, consider a scenario where an elderly person opens an email with the subject line "Receive a free gift" and attempts to click on a link within it. The device captures the link and sends it to the server. The server analyzes the link, and if it determines it may be a phishing attempt, it generates a warning message saying, "This link is suspicious. Do not access it," and sends it to the device. The device displays this warning as a pop-up to the user, allowing the user to refrain from clicking the link.

[0122] An example of a prompt to input into the generation AI model would be something like, "Determine whether the link an elderly person is about to click is safe, and generate a warning message if necessary."

[0123] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0124] Step 1:

[0125] Users use the internet.

[0126] Users open emails or browse websites. For example, a user might try to click a link in an email. This action constitutes input.

[0127] Step 2:

[0128] The device monitors the user's actions.

[0129] The terminal uses monitoring software to monitor user actions in real time. Specifically, it captures the URLs of links the user attempts to click and websites they attempt to access. These URLs are then sent to the server as output.

[0130] Step 3:

[0131] The server detects suspicious links and information.

[0132] The server receives the URL sent from the terminal and uses analysis tools to compare it against a database of known suspicious links and phishing sites. This comparison with the database is data processing, and the server evaluates the safety of the URL. The result of this evaluation is the output.

[0133] Step 4:

[0134] The server generates a warning message.

[0135] Based on the analysis results, the server generates a warning message if it detects suspicious links or information. For example, it might create a message such as, "This link is suspicious. Please stop accessing it." This message is then sent to the terminal as output.

[0136] Step 5:

[0137] The device displays a warning to the user.

[0138] The device receives warning messages sent from the server and displays them to the user. Specifically, these are displayed as pop-up notifications or on-screen alerts. This display is the output, alerting the user to the warning. The user can then review the warning and choose a safe course of action.

[0139] (Application Example 2)

[0140] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0141] When elderly people use the internet, there is an increasing risk of fraud and security risks due to accessing suspicious links or information. In this situation, there is a need to provide an environment where elderly people can use the internet with peace of mind.

[0142] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0143] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for issuing warnings and advice regarding the detected suspicious information, a means for scanning links in real time in cooperation with a database for evaluating the safety of links, and a means for displaying warnings based on the evaluation results. This prevents elderly people from accessing suspicious links and information, enabling them to use the internet safely.

[0144] "Elderly people" refers to individuals whose physical or cognitive abilities are likely to decline due to aging.

[0145] "Suspicious communication" refers to means of information transmission, such as phone calls or emails, that contain content that the recipient finds questionable.

[0146] "Generative artificial intelligence methods" refer to technologies that use machine learning and natural language processing to analyze data and make judgments and predictions according to specific purposes.

[0147] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[0148] "Means for detecting suspicious information" refers to technologies for identifying and distinguishing suspicious elements from received information.

[0149] "Means of providing warnings and advice" refers to notification methods that provide users with warnings and guidance on how to act.

[0150] A "database for evaluating link safety" refers to a collection of information used to assess the reliability and safety of links on the internet.

[0151] "Methods for scanning links in real time" refer to technologies that instantly assess the security of a link the moment it is clicked.

[0152] "Means of displaying warnings based on evaluation results" refers to methods for displaying appropriate warnings to users in accordance with the results of link safety evaluations.

[0153] The system for implementing this invention mainly consists of a server and a user terminal. The server uses generative artificial intelligence to analyze suspicious communications received by elderly people and determine if they are fraudulent. Specifically, the server utilizes natural language processing technology to analyze the content of received emails and phone calls. This allows it to identify potentially fraudulent information and issue warnings to the user.

[0154] The user's device has an application installed to evaluate the security of links. This application scans links in real time as they are clicked and evaluates their security in conjunction with a database. Based on the evaluation results, the device displays a visual or audible warning to the user. This helps prevent elderly people from accessing suspicious links.

[0155] For example, if a user attempts to click a link in an email, and the system determines that the link may lead to a phishing site, it will display a warning message such as, "This link is not secure. Please stop accessing it." An example of a prompt message generated using an AI model would be, "Is this link secure? Please evaluate the link's URL."

[0156] In this way, the server and the user's terminal work together to provide an environment that allows elderly people to use the internet safely.

[0157] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0158] Step 1:

[0159] The user receives an email. The email content is displayed on the device. The user attempts to click a link in the email.

[0160] Step 2:

[0161] The device retrieves the URL of the clicked link. It then sends the retrieved URL to the server.

[0162] Step 3:

[0163] The server analyzes the received URL using a generation AI model. Specifically, it uses natural language processing technology to evaluate the structure and content of the URL and determine its potential for fraud. As a result of the analysis, it outputs an evaluation of whether the URL is safe or not.

[0164] Step 4:

[0165] The server sends the URL evaluation result to the device. The evaluation result includes information on whether the link is safe or suspicious.

[0166] Step 5:

[0167] The device displays a warning to the user based on the evaluation results received from the server. Specifically, if a link is deemed suspicious, it displays a warning message such as, "This link is not secure. Please stop accessing it."

[0168] Step 6:

[0169] The user reviews the warning from their device and decides whether to cancel access to the link. If the user cancels access, the device blocks the connection to the link.

[0170] This series of processes prevents elderly people from accessing suspicious links, enabling them to use the internet safely.

[0171] (Example 3)

[0172] Next, we will describe Embodiment 3 of Embodiment Example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0173] There is a need to provide effective warnings and advice to prevent elderly people from becoming victims of fraud by addressing suspicious communications and links they receive. However, conventional methods are insufficient in detecting suspicious links and providing warnings, and the risk of elderly people becoming victims of fraud still remains. To solve this problem, notifications in a format that elderly people can easily understand and more accurate detection of suspicious links are necessary.

[0174] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0175] In this invention, the server includes generative artificial intelligence means for analyzing suspicious communications received by elderly people, means for determining the possibility of fraud, and means for analyzing suspicious links and evaluating their safety using a machine learning model. This makes it possible to provide elderly people with visually or audibly effective warnings and advice, thereby preventing them from becoming victims of fraud.

[0176] "Generative artificial intelligence means" refers to artificial intelligence technology used to analyze suspicious communications received by elderly people and determine the possibility of fraud.

[0177] "Means of determining the possibility of fraud" refers to the process of analyzing the content of received communications or links and evaluating whether they are related to fraudulent activity.

[0178] "Means for detecting suspicious links and information" refers to technologies that identify suspicious elements within communications and links and issue warnings to users.

[0179] "Means of providing warnings and advice" refers to methods of providing users with messages or guidelines to alert them to suspicious links or information that have been detected.

[0180] "Methods for evaluating safety using machine learning models" refers to the process of analyzing links and information using algorithms that have learned from past data in order to evaluate their safety.

[0181] "Means of providing visual or auditory notification" refers to methods of communicating warnings about suspicious links or information to users through visual displays or audio.

[0182] As an embodiment of this invention, the following system is constructed.

[0183] The server analyzes suspicious communications received by elderly individuals using a generative AI model. This involves the use of natural language processing libraries and machine learning frameworks. Specifically, it utilizes software such as "spaCy" and "TENSORFLOW®" to analyze communication content and detect suspicious links and information. The server then uses a machine learning model to evaluate the safety of detected suspicious links and determine their potential for fraud.

[0184] The device receives warning messages and advice sent from the server and notifies the elderly visually or audibly. Specifically, it uses pop-up messages and audio alerts to warn users not to click on suspicious links.

[0185] The user can initiate the system's operation by entering prompts for the generated AI model. For example, the following prompts might be used:

[0186] "Please create a program that can determine if links in emails received by elderly people are safe and display a warning if necessary."

[0187] Based on this prompt, the generating AI model creates a program to detect suspicious links and issue warnings. This enables elderly people to take concrete actions to prevent themselves from becoming victims of fraud. The flow of the specific processing in Example 3 will be explained using Figure 15.

[0188] Step 1:

[0189] The user inputs a prompt into the generative AI model. For example, the prompt might be, "Create a program that determines whether a link in an email received by an elderly person is safe and displays a warning if necessary." Based on this input, the generative AI model generates a program to detect suspicious links.

[0190] Step 2:

[0191] The server uses the generated program to analyze emails and communications received by elderly individuals. The input includes email content and link information. The server uses the natural language processing library "spaCy" to analyze the email content and extract links and domain information. The output provides potential suspicious links.

[0192] Step 3:

[0193] The server evaluates the extracted link information using a machine learning model. The input includes the suspicious link candidates obtained in step 2. The server uses TensorFlow to assess the safety of the links and determine their likelihood of being fraudulent. The output is a list of suspicious links.

[0194] Step 4:

[0195] The server generates warning messages and advice for links it deems suspicious. The input includes the list of suspicious links obtained in step 3. The server creates a warning message for the user, such as "This link is suspicious. Do not click it." The output is the generated warning message.

[0196] Step 5:

[0197] The server sends the generated warning message to the terminal. The input includes the warning message generated in step 4. The server prepares to send the warning message to the terminal and notify the user. The output is that the message has been sent to the terminal.

[0198] Step 6:

[0199] The terminal displays warning messages received from the server to the user. Input includes the warning messages sent from the server. The terminal provides visual or auditory notifications to the elderly using pop-up messages or audio alerts. Output is the completion of the warning to the user.

[0200] (Application Example 3)

[0201] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0202] One challenge when elderly people use digital communications is preventing them from becoming victims of fraud or malicious links. In particular, elderly people have difficulty distinguishing suspicious links and information, which increases their risk of becoming victims of fraud. There is a need to improve this situation and provide an environment where elderly people can use digital communications with peace of mind.

[0203] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0204] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for scanning communication links in real time and issuing a warning when a suspicious link is detected, and a means for issuing warnings through visual notifications or audio alerts. This makes it possible for elderly people to easily identify suspicious links and information and prevent themselves from becoming victims of fraud.

[0205] "Elderly people" generally refers to older individuals who require support in using digital technologies.

[0206] "Suspicious communications" refers to means of information transmission such as phone calls, emails, and messages that may be involved in fraud or illegal activities.

[0207] "Generative artificial intelligence" refers to artificial intelligence technologies used to analyze data and detect specific patterns or anomalies.

[0208] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[0209] "Means for detecting suspicious information" refers to technologies for identifying fraudulent or questionable elements from received information.

[0210] "Real-time scanning" refers to a technology that analyzes information instantly upon receipt and provides the results.

[0211] "Warning mechanisms" refer to technologies that provide notifications and alerts to inform users of potential dangers.

[0212] "Visual notifications" refer to means of conveying information through visual means, such as messages and pop-ups displayed on the screen.

[0213] "Voice alert" refers to a warning method that uses sound to draw the user's attention.

[0214] The system for implementing this invention mainly consists of a server and a terminal. The server uses generative artificial intelligence to analyze suspicious communications received by elderly people and determine if they are fraudulent. Specifically, the server utilizes natural language processing technology to analyze the content of communications. This allows it to detect suspicious information and perform real-time scans.

[0215] The device scans the links of communications received by the user in real time and issues a warning if it detects a suspicious link. The warning is provided to the user through visual notifications and audio alerts. Visual notifications appear as pop-ups on the device screen, and audio alerts are generated using a Text-to-Speech (TTS) engine.

[0216] As a concrete example, a device automatically scans a link received by a user via SMS and displays a warning message stating, "This link may lead to a phishing site. Do not click it." In this process, the server uses the Google® Safe Browsing API to evaluate the safety of the link.

[0217] An example of a prompt to input into a generating AI model is, "Is this link safe? If it may be a phishing site, please display a warning." Using this prompt, the system can provide appropriate warnings to the user.

[0218] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[0219] Step 1:

[0220] The user receives the communication on their device. The device extracts links and text data to analyze the content of the received communication. The input is the received communication data, and the output is the links and text data to be analyzed.

[0221] Step 2:

[0222] The device sends extracted links to the server for real-time scanning. The server uses the Google Safe Browsing API to evaluate the safety of the links. The input is the links to be analyzed, and the output is the evaluation result regarding the safety of the links.

[0223] Step 3:

[0224] The server uses generative artificial intelligence to analyze the text data of received communications using natural language processing. This allows it to determine whether the communications are potentially fraudulent. The input is the text data of the communications, and the output is the evaluation result regarding the likelihood of fraud.

[0225] Step 4:

[0226] The server integrates the link safety assessment results and the text data analysis results, and generates a warning if suspicious information is detected. The input is the link safety assessment results and the fraud potential assessment results, and the output is a warning message.

[0227] Step 5:

[0228] The terminal notifies the user of warning messages received from the server. It displays a pop-up on the screen as a visual notification and plays the warning as an audio alert using a Text-to-Speech (TTS) engine. The input is the warning message, and the output is the warning notification to the user.

[0229] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0230] "Example of form 1"

[0231] One embodiment of the present invention involves a system that combines a generative AI for analyzing suspicious phone calls and emails received by elderly people, means for determining the possibility of fraud, means for detecting suspicious links and information provided by elderly people, and means for providing warnings and advice regarding detected suspicious links and information, with an emotion engine. This emotion engine analyzes the emotions of elderly people from their tone of voice and facial expressions, and incorporates the results as part of the fraud judgment. Specifically, it analyzes the tone and speed of the elderly person's voice when they speak on the phone, or changes in facial expressions read from video calls or photographs, and reflects the results in the fraud judgment.

[0232] "Example of form 2"

[0233] Furthermore, the emotional engine adjusts how warnings and advice are delivered based on the emotions of the elderly. For example, if it determines that an elderly person is feeling anger or confusion, it uses that information to adjust the tone and expression of warnings and advice, providing information in a way that is easy for the elderly person to understand. This makes it possible to provide appropriate advice and protect elderly people from fraud even when they are emotionally charged.

[0234] "Example of form 3"

[0235] Furthermore, the emotion engine can track changes in the emotions of older adults over the long term and incorporate that information into fraud detection. For example, if an older adult is consistently judged to be experiencing anxiety or fear, this information can be used to increase the likelihood of fraud and encourage more careful monitoring. This allows for the use of changes in the emotions of older adults in fraud detection, resulting in more accurate fraud detection.

[0236] The following describes the processing flow for each example of the form.

[0237] "Example of form 1"

[0238] Step 1: A generative AI analyzes suspicious phone calls and emails received by elderly individuals.

[0239] Step 2: Based on the analysis results, determine the likelihood of fraud.

[0240] Step 3: Detect suspicious links and information provided by elderly individuals.

[0241] Step 4: The emotion engine analyzes the emotions of the elderly person from their tone of voice and facial expressions.

[0242] Step 5: Use the analysis results as part of the fraud determination.

[0243] "Example of form 2"

[0244] Step 1: The emotion engine analyzes the emotions of the elderly.

[0245] Step 2: Based on the analysis results, adjust how warnings and advice are provided.

[0246] Step 3: Provide adjusted warnings and advice to the elderly.

[0247] "Example of form 3"

[0248] Step 1: The emotion engine tracks changes in older adults' emotions over the long term.

[0249] Step 2: Use the tracking results to determine if it's a scam.

[0250] Step 3: Based on the results, increase the likelihood of fraud and monitor more carefully.

[0251] (Example 1)

[0252] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0253] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to provide appropriate warnings and advice. However, conventional technologies are insufficient in identifying fraud or may delay warnings, which presents a challenge as the elderly are at a high risk of becoming victims of fraud.

[0254] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0255] In this invention, the server includes information processing means, evaluation means, identification means, sentiment analysis means, and notification means. This makes it possible to quickly analyze suspicious communications received by the elderly, determine the possibility of fraud with high accuracy, and provide appropriate warnings and advice.

[0256] "Information processing means" refers to technical means for analyzing suspicious communications received by elderly people, and has the function of analyzing the content of communications using natural language processing technology.

[0257] "Evaluation means" refers to technical means for determining the possibility of fraud based on the analyzed communication content, and has the function of scoring the possibility of fraud using a generative AI model.

[0258] "Identification means" refers to technical means for detecting suspicious information provided by elderly individuals, and includes the function of evaluating the security of links and information.

[0259] "Emotional analysis means" refers to technical means for analyzing the emotions of elderly people and reflecting the results in fraud detection. These means have the function of analyzing emotions using voice recognition technology and image analysis technology.

[0260] "Notification means" refers to technical means for issuing warnings or advice regarding detected suspicious information, and has the function of providing visual or auditory notifications to the elderly.

[0261] This invention is a system for analyzing suspicious communications received by elderly people and determining the possibility of fraud. The system includes information processing means, evaluation means, identification means, sentiment analysis means, and notification means.

[0262] The server uses software such as Python and TensorFlow as information processing tools to analyze the content of suspicious communications received by elderly individuals using natural language processing techniques. This makes it possible to extract keywords and phrases contained in the communications and analyze the context.

[0263] As an evaluation method, the server uses a generative AI model to score the likelihood of fraud based on the analyzed data. If it is determined that there is a high probability of fraud, the process proceeds to the next step.

[0264] The identification mechanism works to detect suspicious links and information provided by users. The server evaluates the safety of the link and issues a warning if it determines that it is dangerous.

[0265] As a means of emotion analysis, the server uses speech recognition and image analysis technologies to analyze the tone and speed of the elderly person's voice, as well as changes in their facial expressions. This allows the server to determine whether the elderly person is feeling anxious or tense, and this information is then used to determine if they are a fraud suspect.

[0266] As a notification method, the server sends warnings and advice to the terminal. The terminal provides elderly users with visual or auditory notifications and displays specific advice. For example, it may display a message such as, "This call may be a scam. Do not provide any personal information."

[0267] For example, if a user receives a suspicious email, their device forwards the email to the server. The server analyzes the email's content, and if it detects phrases such as "verify bank account information," it warns the user, "This email may be a scam. Do not click the link."

[0268] Examples of prompt messages include the following:

[0269] "Could this phone call be a scam?"

[0270] "Are the links in the email safe?"

[0271] "Does this tone of voice indicate anxiety?"

[0272] In this way, the system analyzes suspicious information received by the elderly, determines the possibility of fraud, and issues a warning if necessary.

[0273] The flow of the specific processing in Example 1 will be explained using Figure 17.

[0274] Step 1:

[0275] When a user receives a suspicious phone call or email, they input the content into their device. The device then sends the audio or text data to the server. Specifically, data is collected using the smartphone's recording function or email app. The input is audio files or text data, and the output is data transfer to the server.

[0276] Step 2:

[0277] The server analyzes the received data using natural language processing technology. Using Python and TensorFlow, it extracts keywords and phrases contained in the content of calls and emails, and analyzes the context. The inputs are audio files and text data, and the output is a list of analyzed keywords and phrases.

[0278] Step 3:

[0279] The server determines the possibility of fraud based on the analysis result. Using a generative AI model, it scores the possibility of fraud from the combination of detected keywords and phrases. The input is the list of analyzed keywords and phrases, and the output is a score indicating the possibility of fraud.

[0280] Step 4:

[0281] The server analyzes the user's emotion using an emotion engine. Using speech recognition technology, it analyzes the tone and speed of the user's voice, and using image analysis technology, it reads changes in facial expressions from video calls and photos. The inputs are audio data and image data, and the output is the result of emotion analysis.

[0282] Step 5:

[0283] The server sends a warning to the user when it is determined that the possibility of fraud is high. It sends a notification to the terminal and displays specific advice and points for attention. The inputs are the score indicating the possibility of fraud and the result of emotion analysis, and the output is a warning message to the user. As a specific operation, a message such as "This call may be a fraud. Please do not provide personal information." is displayed.

[0284] (Application Example 1)

[0285] Next, Application Example 1 of Embodiment 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0286] There is a need to quickly and accurately determine the possibility of fraud in suspicious communications received by the elderly and to provide appropriate warnings and advice. However, conventional technologies have limitations in analyzing the content of communications and do not take into account the emotional state of the elderly when making fraud judgments, making it difficult to prevent fraud.

[0287] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0288] In this invention, the server includes a generative information processing means for analyzing suspicious communications received by an elderly person, an information processing means for determining the possibility of fraud, and an information processing means for analyzing the elderly person's voice and facial expressions to detect emotions. This makes it possible to make fraud judgments that take into account the emotional state of the elderly person.

[0289] "Elderly people" refers to individuals whose physical or cognitive abilities are likely to decline as a result of aging.

[0290] "Suspicious communication" refers to means of transmitting information, such as phone calls or emails, that may have questionable intentions or content directed at the recipient.

[0291] "Generative information processing means" refers to information processing technology that analyzes received information and makes decisions or generates information according to a specific purpose.

[0292] "Information processing means for determining the possibility of fraud" refers to information processing technology that analyzes received information and evaluates whether that information is related to fraudulent activity.

[0293] "Information processing means for detecting suspicious information" refers to information processing technology for identifying and detecting suspicious elements from received information.

[0294] "Information processing means for issuing warnings and advice" refers to information processing technology that provides recipients with warnings and guidance on action based on detected suspicious information.

[0295] "Information processing means for analyzing voice and facial expressions to detect emotions" refers to information processing technology that analyzes the voice and facial expressions of a recipient to identify their emotional state.

[0296] "Information processing means that reflect the results of emotional analysis in fraud judgment" refers to information processing technology that incorporates analyzed emotional information into the process of evaluating the likelihood of fraud.

[0297] The system for implementing this invention mainly consists of a server and a terminal. The server is equipped with generative information processing means for analyzing suspicious communications received by elderly people, and analyzes the content of the communications using natural language processing technology. Specifically, it uses natural language processing libraries (e.g., spaCy or NLTK) to extract keywords and phrases from the communications and evaluate the possibility of fraud.

[0298] Furthermore, the server is equipped with information processing capabilities to analyze the voice and facial expressions of elderly individuals and detect their emotions. This involves using emotion analysis libraries (such as OpenCV or DeepFace) to analyze audio and video data acquired through the device's camera and microphone. The analysis results are used to determine if a fraud is likely, and if there is a high probability of fraud, warnings and advice are displayed on the device.

[0299] The device sends communications received by the user to a server and receives analysis results from the server. The user receives visual or auditory notifications on the device to be alerted to potentially fraudulent communications.

[0300] For example, if a user receives an email titled "Important Notice from the Bank," the device sends its contents to a server. The server uses natural language processing to analyze keywords such as "bank," "important," and "notice" to determine if it is a scam. Additionally, when a user receives a phone call, the device's camera analyzes their facial expressions to detect changes in emotion. If it is determined that the email is likely to be a scam, a warning message such as "This email may be a scam. Do not click the link." is displayed on the device.

[0301] An example of a prompt text input to a generative AI model includes: "Analyze the content of this email and determine the possibility of fraud. The content of the email is as follows: 'This is an important notice from the bank.'"

[0302] The flow of the specific processing in Application Example 1 will be described using FIG. 18.

[0303] Step 1:

[0304] The terminal transmits the suspicious communication (call or email) received by the user to the server. The input is the content of the communication, and the output is the transmission data to the server. The terminal converts the communication content into text format, and generates a data packet for transmission to the server.

[0305] Step 2:

[0306] The server analyzes the received communication content using the generative information processing unit. The input is the communication content transmitted from the terminal, and the output is the analysis result. The server uses a natural language processing library (for example, spaCy or NLTK) to extract keywords and phrases from the communication content, and evaluate the possibility of fraud.

[0307] Step 3:

[0308] The server analyzes the user's voice and facial expressions to detect emotions. The input is voice and video data transmitted from the terminal, and the output is the emotion analysis result. The server uses an emotion analysis library (for example, OpenCV or DeepFace) to analyze the tone of the voice and changes in facial expressions.

[0309] Step 4:

[0310] The server integrates the results of communication content analysis and sentiment analysis to determine the likelihood of fraud. The input is the results of communication content analysis and sentiment analysis, and the output is the evaluation of the likelihood of fraud. The server uses a generative AI model to make a comprehensive judgment on the likelihood of fraud.

[0311] Step 5:

[0312] The server sends a warning or advice to the terminal if it determines that there is a high probability of fraud. The input is the result of the fraud probability assessment, and the output is the warning message. The server generates the warning message and creates a data packet to send to the terminal.

[0313] Step 6:

[0314] The terminal notifies the user of warnings and advice received from the server. Input is the warning message from the server, and output is the notification to the user. The terminal displays or plays the warning message to the user visually or audibly.

[0315] (Example 2)

[0316] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0317] When elderly people use the internet, there is a problem in that they are at a higher risk of becoming victims of fraud because they cannot respond appropriately to suspicious communications and information. In addition, they may not be able to correctly understand information because they are not provided with appropriate warnings and advice that are tailored to their emotional state.

[0318] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0319] In this invention, the server includes a generative AI means for analyzing suspicious communications received by elderly people, means for determining the possibility of fraud, and means for detecting suspicious information provided by elderly people. This enables elderly people to respond quickly and appropriately to suspicious information and prevent them from becoming victims of fraud.

[0320] "Generative AI methods" refer to artificial intelligence technologies used to analyze suspicious communications received by elderly people and to understand their content.

[0321] "Means for determining the possibility of fraud" refers to algorithms or processes that evaluate whether a communication is fraudulent based on its analyzed content.

[0322] "Methods for detecting suspicious information" refer to technologies for identifying information that exhibits unusual patterns or has known risks from information provided by elderly individuals.

[0323] "Warning measures" refer to methods of providing visual or auditory notifications to alert older adults to detected suspicious information.

[0324] "Methods for analyzing emotional states" refer to techniques for analyzing the tone of voice and facial expressions of elderly people to determine their emotional state.

[0325] "Means of adjusting the content of warnings" refers to methods of modifying the tone and expression of warnings and advice to make them easier for older adults to understand, based on their analyzed emotional state.

[0326] This invention is a system that enables elderly people to appropriately respond to suspicious communications and information when using the internet. The system consists of three components: a server, a terminal, and a user.

[0327] The server uses generative AI methods to analyze suspicious communications received by elderly individuals. This analysis employs natural language processing techniques to understand the content of the communications and determine if they are fraudulent. The server runs algorithms that cross-reference information with a database to identify information that poses a known risk.

[0328] The device uses voice recognition software and a facial recognition camera to analyze the emotional state of elderly individuals. This allows the device to analyze the tone of voice and facial expressions of elderly individuals in real time to determine their emotional state. Based on this emotional state, the device adjusts the content of warnings and advice, providing information in a way that is easy for the elderly to understand.

[0329] Elderly users receive warnings and advice from the device and take appropriate action. For example, when an elderly person inputs the prompt "Is it safe to click this link?" into the AI ​​model, the system evaluates the safety of the link and returns appropriate advice. This allows elderly people to use the internet with peace of mind.

[0330] The flow of the specific processing in Example 2 will be explained using Figure 19.

[0331] Step 1:

[0332] The server monitors communications received by elderly individuals and retrieves emails and website links as input. Using generative AI, the server analyzes the content of these communications using natural language processing techniques. This analysis generates data to determine the likelihood of a communication being fraudulent. The output provides the analysis results and a flag indicating whether the communication is potentially fraudulent.

[0333] Step 2:

[0334] The server detects suspicious information based on the analysis results. It uses a database of analyzed communication content and information with known risks as input. The server compares these and runs algorithms to identify suspicious links and information. A list of suspicious information is generated as output.

[0335] Step 3:

[0336] The terminal receives a list of suspicious information sent from the server and uses it as input. The terminal uses an emotion engine to analyze the emotional state of elderly people. Specifically, it uses speech recognition software and a facial recognition camera to analyze the tone of voice and facial expressions of elderly people in real time. The output is an evaluation result of their emotional state.

[0337] Step 4:

[0338] The device adjusts the content of warnings and advice based on the results of an emotional state assessment. It uses a list of suspicious information and the results of the emotional state assessment as input. The device modifies the tone and expression of warnings and advice to make them easily understandable to older adults. The adjusted warnings and advice are generated as output and provided to older adults as visual or auditory notifications.

[0339] Step 5:

[0340] The elderly user receives warnings and advice from the device and takes appropriate action. Specifically, the elderly user can input a prompt message such as "Is it safe to click this link?" into the AI ​​model. Based on this prompt, the system evaluates the safety of the link and provides feedback to the user. As output, the elderly user is presented with advice regarding the safety of the link.

[0341] (Application Example 2)

[0342] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0343] Fraud cases involving elderly people accessing suspicious links and information online are on the rise. In this situation, there is a need to provide an environment where elderly people can use the internet with peace of mind. Furthermore, it is important to provide appropriate warnings and advice that take into account the feelings of the elderly.

[0344] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0345] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for issuing warnings and advice regarding the detected suspicious information, an emotion analysis means for analyzing the emotions of elderly people and adjusting the expression of warnings and advice, and a means for verifying the safety of suspicious information using an external database. This makes it possible to reduce the risk of elderly people accessing suspicious information and to provide appropriate warnings and advice that take their emotions into consideration.

[0346] "Elderly people" refers to individuals who are considered to be prone to a decline in physical or cognitive abilities as a result of aging.

[0347] "Suspicious communication" refers to means of information transmission such as phone calls and emails that may have been sent to the recipient with an improper purpose.

[0348] "Generative artificial intelligence means" refers to methods that use artificial intelligence technology to analyze data and make judgments or predictions according to specific purposes.

[0349] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[0350] "Means of detecting suspicious information" refers to methods for identifying information that is deemed unreliable or dangerous from among the information provided on the internet.

[0351] "Means of providing warnings and advice" refers to means of generating notifications and messages to draw attention to users or provide guidance on how to act.

[0352] "Emotional analysis tools" refer to methods for analyzing a user's emotional state and taking appropriate action based on the results.

[0353] "Methods of verification using external databases" refer to methods of verifying information by referring to external databases in order to evaluate the safety and reliability of information on the internet.

[0354] The system for implementing this invention mainly consists of a server and a user terminal. The server uses generative artificial intelligence means to analyze suspicious communications received by elderly people. Specifically, a program developed using Python utilizes the Google Cloud Natural Language API to analyze the content of the communications using natural language processing. This determines whether the communications are likely to be fraudulent.

[0355] On the user's device, it is implemented as an internet browser extension to detect suspicious information. It uses the VirusTotal API to verify the safety of links and information. For detected suspicious information, a notification is displayed to provide warnings and advice. This notification is delivered in an appropriate tone, based on an analysis of the user's emotional state using sentiment analysis tools. The Google Cloud Natural Language API is used for sentiment analysis.

[0356] For example, if an elderly person attempts to click on a link in an email, the system may detect that the link is suspicious. If the sentiment analysis system determines that the user is confused, a gentle warning message will be displayed, such as, "This link may not be safe. Please check the details."

[0357] An example of a prompt message is, "The user is about to click a link. Please verify the safety of this link and display a warning appropriate to the user's sentiment." This prompt message allows the system to take appropriate action and protect elderly users from suspicious information.

[0358] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[0359] Step 1:

[0360] The user attempts to click a link on their device. The device retrieves the URL of the clicked link. This URL becomes the input data.

[0361] Step 2:

[0362] The device sends the retrieved URL to the VirusTotal API for security verification. The VirusTotal API evaluates the URL's security and outputs the result. The output indicates whether the URL is safe or suspicious.

[0363] Step 3:

[0364] The server receives evaluation results from the VirusTotal API, and if it determines that the URL is suspicious, it collects sentiment data from the user's device to analyze the user's sentiment. This data includes the user's recent activity history and input content.

[0365] Step 4:

[0366] The server sends the collected emotion data to the Google Cloud Natural Language API for emotion analysis. The input data is information about the user's emotions, and the output is the analysis result showing the user's emotional state.

[0367] Step 5:

[0368] The server generates a warning message in an appropriate tone based on the sentiment analysis results. It uses a generative AI model to create prompts that respond to the user's emotions. For example, if the user is confused, it might generate a message such as, "This link may not be safe. Please check the details."

[0369] Step 6:

[0370] The device displays warning messages received from the server to the user. This allows the user to receive warnings about suspicious links and take appropriate action.

[0371] (Example 3)

[0372] Next, we will describe Embodiment 3 of Embodiment Example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0373] Reducing the risk of elderly people becoming victims of fraud is crucial. In particular, it is important to appropriately identify suspicious communications and information and issue prompt warnings. Improving the accuracy of fraud detection, taking into account the emotional changes of the elderly, is also a challenge.

[0374] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0375] In this invention, the server includes information processing means, judgment means, detection means, notification means, and emotion analysis means. This makes it possible to analyze suspicious communications received by elderly people, assess the possibility of fraud, and provide appropriate warnings and advice. Furthermore, by tracking changes in the elderly person's emotions and reflecting them in fraud judgments, the accuracy of fraud detection can be improved.

[0376] "Information processing means" refers to technology for analyzing suspicious communications received by elderly people and understanding their content.

[0377] "Decision-making tools" are techniques used to evaluate the likelihood of fraud based on analyzed information.

[0378] "Detection means" refers to technology for identifying and distinguishing suspicious information provided by elderly individuals.

[0379] "Notification methods" refer to technologies used to communicate warnings and advice to the elderly regarding detected suspicious information.

[0380] "Emotional analysis methods" refer to technologies that track changes in the emotions of elderly people and use that information to make fraud judgments.

[0381] This invention is a fraud detection system for the elderly, realized through the respective roles of the server, terminal, and user. The server uses information processing means, judgment means, detection means, notification means, and sentiment analysis means to analyze suspicious communications received by the elderly and evaluate the possibility of fraud.

[0382] Specifically, the server uses a generative AI model to scan links and information contained in emails and messages to detect suspicious elements. For example, it analyzes the URL of a link and compares it against a list of known fraudulent sites to determine the likelihood of fraud. It also uses an emotion engine to collect user emotion data and track changes in their emotions. This allows the system to incorporate information about when a user is feeling anxious or fearful into its fraud detection.

[0383] The device provides visual or auditory notifications to the elderly based on instructions from the server. For example, if a user attempts to click a suspicious link, the device displays a warning such as, "This link may be a scam. Do not click it." It also provides advice based on emotional data, such as, "You seem to have been feeling anxious recently. Be careful of scams."

[0384] Users receive notifications from their devices and take specific actions. For example, they may be prompted not to click on a link they are warned about or to delete a suspicious email.

[0385] An example of a prompt message is, "Please determine if the link in this email is safe." The generating AI model then analyzes the link's safety and notifies the user of the result. Figure 21 illustrates the specific processing flow in Example 3.

[0386] Step 1:

[0387] Users generate data through everyday actions. Inputs include receiving emails, browsing websites, and sending and receiving messages. This data is sent to a server. The server collects this data and prepares it for analysis. Specifically, it stores email content and link information in a database.

[0388] Step 2:

[0389] The server analyzes the collected data. The input is data from emails and messages sent by users. The server uses a generative AI model to scan links and information contained in emails and messages, detecting suspicious elements. Specifically, it analyzes the URLs of links and compares them against a list of known fraudulent websites. The output is a list of suspicious links and information.

[0390] Step 3:

[0391] The server uses an emotion engine to collect user emotion data. Input includes user behavior and input data while using the device. The server analyzes this data to infer changes in the user's emotions. Specifically, if the user frequently exhibits anxiety-inducing behaviors, this data is recorded. The output is a report on changes in emotions.

[0392] Step 4:

[0393] The server integrates analysis results and sentiment data to determine the likelihood of fraud. Inputs include a list of suspicious links and information, and reports on changes in sentiment. Based on this data, the server assesses the risk of fraud and generates warnings as needed. The output is a warning message.

[0394] Step 5:

[0395] The device provides visual or auditory notifications to the user based on instructions from the server. The input is a warning message from the server. If the user attempts to click a suspicious link, the device will display a warning such as, "This link may be fraudulent. Do not click it." Specifically, this involves displaying a large red warning message on the screen. The output is a warning notification to the user.

[0396] Step 6:

[0397] The user receives a notification from their device and takes specific action. The input is a warning notification from the device. The user is prompted to take action such as not clicking the warned link or deleting a suspicious email. The specific action is to choose an action to avoid the risk of fraud. The output is that the user's safe actions are promoted.

[0398] (Application Example 3)

[0399] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0400] There is a need to reduce the risk of elderly people becoming victims of fraud. However, conventional methods have difficulty in appropriately detecting and warning about suspicious communications and information, and fraud judgments do not take into account the emotional changes of the elderly, resulting in insufficient accuracy in fraud prevention.

[0401] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0402] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, and an emotion analysis means for tracking changes in the elderly person's emotions and reflecting them in the fraud determination. This reduces the risk of fraud against elderly people and enables more accurate fraud prevention.

[0403] "Generative artificial intelligence methods" refer to methods that use artificial intelligence technology to analyze suspicious communications received by elderly people and determine the possibility of fraud.

[0404] "Means of determining the possibility of fraud" refers to methods of evaluating whether the analyzed information is related to fraud.

[0405] "Means for detecting suspicious information" refers to methods for identifying suspicious elements in information provided by elderly individuals and issuing warnings.

[0406] "Means of providing warnings or advice" refers to means of visually or audibly drawing the attention of elderly individuals to any suspicious information that has been detected.

[0407] "Emotional analysis methods" refer to techniques for tracking changes in the emotions of elderly individuals and using that information to inform fraud detection.

[0408] "Notification means" refers to methods for sending visual or auditory warnings to elderly people.

[0409] The system for implementing this invention mainly consists of a server and a terminal. The server is equipped with generative artificial intelligence means for analyzing suspicious communications received by elderly people. This means analyzes the content of communications using natural language processing technology and determines the possibility of fraud. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to build a model for analyzing the content of communications.

[0410] The devices used by the elderly are smartphones, tablets, and other devices on which notification mechanisms are implemented. These notification mechanisms issue visual or auditory warnings based on analysis results. For example, visual warnings can be displayed using ANDROID® Notification or iOS Push Notification, and voice warnings can be issued using Google Assistant or Siri.

[0411] Furthermore, the server is equipped with emotion analysis capabilities to track changes in the emotions of elderly individuals. This capability collects voice and facial expression data from elderly individuals and uses an AI model to analyze changes in their emotions. This makes it possible to incorporate emotional information into fraud detection.

[0412] For example, if an elderly person opens an email on their smartphone and it contains a suspicious link, the device will issue a voice warning saying, "This link may be a scam. Do not click it," and display a red warning message on the screen.

[0413] An example of a prompt message would be, "Is this link safe? If it may be a scam, please generate a warning message." This would enable older adults to take concrete actions to protect themselves from fraud.

[0414] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[0415] Step 1:

[0416] The server receives communication data received by elderly individuals. The input consists of email and phone call content, which is then passed to a generative artificial intelligence system. The server analyzes the communication content using natural language processing techniques to determine the likelihood of fraud. The output is the result of the determination of whether or not fraud is likely.

[0417] Step 2:

[0418] The server detects suspicious information based on the analysis results. The input is the judgment result obtained in step 1, which is used to identify suspicious links and information. The server lists the identified suspicious information and sends it to the terminal. The output is a list of suspicious information.

[0419] Step 3:

[0420] The terminal issues a warning to the user based on a list of suspicious information received from the server. The input is a list of suspicious information, which is used to provide visual or auditory warnings. The terminal displays a warning message on the screen and uses a voice assistant to draw attention. The output is the notification of the warning to the user.

[0421] Step 4:

[0422] The server collects emotional data from elderly individuals and analyzes it using emotion analysis tools. Inputs include voice and facial expression data, which are then analyzed by an AI model. The server tracks emotional changes and incorporates them into fraud detection. The output is the result of the emotion analysis.

[0423] Step 5:

[0424] The server improves the accuracy of its fraud detection based on the sentiment analysis results. The input consists of the sentiment analysis results and the judgment results from Step 1, which are integrated to assess the final fraud risk. The server sends the assessment results to the terminal and strengthens warnings as needed. The output is the final fraud risk assessment.

[0425] (Other examples)

[0426] Next, other embodiments will be described. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0427] Communications received by the elderly may contain suspicious content that could be fraudulent, increasing their risk of becoming victims of fraud. There is a need for a system that can quickly and accurately detect such potential scams and warn the elderly.

[0428] The identification process performed by the identification processing unit 290 of the data processing device 12 in other embodiments is realized by the following means.

[0429] In this invention, the server includes means for generating prompts to instruct the system to analyze the content of communications using a generative AI model and evaluate the possibility of fraud; means for inputting the generated prompts into the generative AI model and evaluating the possibility of fraud; and means for identifying information provided by an elderly person, which includes means for generating prompts to instruct the system to analyze the information using the generative AI model and determine whether the information is related to fraud. This makes it possible to quickly evaluate the possibility of fraud in communications received by elderly people and issue appropriate warnings.

[0430] A "generative AI model" is a model trained using artificial intelligence technology to perform specific tasks, making predictions and decisions based on the input data.

[0431] A "prompt" is a set of instructions given to a generative AI model to perform a specific task, providing guidelines for the model to generate an appropriate response.

[0432] "Information processing means" refers to hardware or software components for collecting, analyzing, and processing data, and which have the function of manipulating data according to a specific purpose.

[0433] A "decision-making tool" is a component that has the function of evaluating specific conditions based on input data and deriving a result, and it uses a generative AI model to assess the possibility of fraud.

[0434] An "identification means" is a component that has the function of analyzing input information and detecting specific patterns or characteristics, and is used to identify information related to fraud.

[0435] The following describes "modes for carrying out the invention."

[0436] ---

[0437] This invention is a system that analyzes suspicious communications received by the elderly and assesses the possibility of fraud. The system mainly consists of three elements: a server, a terminal, and a user.

[0438] The server first retrieves the communication content received by the elderly person. Specifically, it collects communication data using the APIs of the mail server or messaging application. The retrieved communication content is then preprocessed using natural language processing libraries (e.g., NLTK or spaCy). This preprocessing includes text tokenization, stop word removal, and stemming.

[0439] Next, the server uses a generative AI model (e.g., OpenAI's GPT-3®) to generate a prompt message instructing it to assess the likelihood of fraud. This prompt message is input into the generative AI model and forms the basis for assessing the likelihood of fraud. Example prompt message: "Assess the likelihood that this message is fraudulent."

[0440] The device receives an assessment result regarding the likelihood of fraud sent from the server. Based on the assessment result, the device generates a prompt message instructing the AI ​​model to use again to determine if the information provided by the elderly person is related to fraud. This prompt message is something like, "Please determine if this information is related to fraud."

[0441] Users receive warning messages from their devices and take appropriate action against potentially fraudulent communications. For example, they may refrain from clicking on links or providing personal information.

[0442] This system makes it possible to quickly assess the potential for fraud in communications received by the elderly and issue appropriate warnings.

[0443] The flow of specific processing in other embodiments will be explained using Figure 23.

[0444] Step 1:

[0445] The server retrieves the communication content received by the elderly. Specifically, it collects communication data using APIs from mail servers and messaging applications. The input is the communication data, and the output is text data for preprocessing. The server uses natural language processing libraries (e.g., NLTK and spaCy) to perform preprocessing such as text tokenization, stop word removal, and stemming.

[0446] Step 2:

[0447] The server generates a prompt message for input to a generative AI model based on preprocessed text data. This prompt message is something like, "Evaluate the likelihood that this message is a scam." The input is the preprocessed text data, and the output is the generated prompt message. The server inputs the prompt message into a generative AI model (e.g., OpenAI's GPT-3) and evaluates the likelihood of it being a scam.

[0448] Step 3:

[0449] The terminal receives an assessment result regarding the likelihood of fraud sent from the server. The input is the assessment result from the server, and the output is data used to detect suspicious information. Based on the assessment result, the terminal generates a prompt message instructing the AI ​​model to determine whether the information provided by the elderly person is related to fraud in order to detect suspicious information. This prompt message is something like, "Please determine if this information is related to fraud."

[0450] Step 4:

[0451] The device generates a prompt message instructing the AI ​​model to visually or audibly display a warning message to the elderly. This prompt message might be something like, "Display this message to the elderly as a warning." The input is the evaluation result regarding the likelihood of fraud, and the output is the warning message to be sent to the elderly. The device then notifies the elderly of the generated warning message.

[0452] Step 5:

[0453] The user receives a warning message from their device and takes appropriate action against potentially fraudulent communications. The input is the warning message from the device, and the output is the user's actions. Specifically, this includes actions such as not clicking on links or not providing personal information.

[0454] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0455] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0456] Other examples of generative AI include Gemini® (registered trademark) (Internet search). <url: https: gemini.google.com ?hl="ja">) are some examples.

[0457] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0458] [Second Embodiment]

[0459] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0460] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0461] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0462] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0463] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0464] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0465] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0466] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0467] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0468] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0469] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0470] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0471] "Example of form 1"

[0472] One embodiment of the present invention involves a generative AI system that analyzes suspicious phone calls and emails received by elderly individuals. This generative AI uses natural language processing technology to analyze the content of phone calls and emails and determine the likelihood of them being scams. Specifically, it analyzes keywords, phrases, and context contained in the content of phone calls and emails, and if it determines that there is a high probability of fraud, it warns the elderly individual with that information.

[0473] "Example of form 2"

[0474] Furthermore, there are methods to detect suspicious links and information provided by elderly individuals. These methods monitor the actions of elderly individuals when they use the internet, and if they attempt to provide suspicious links or information, they detect the action and warn the elderly individual. Specifically, they detect actions such as elderly individuals attempting to click on links in emails or access suspicious websites, and warn the elderly individual.

[0475] "Example of form 3"

[0476] Furthermore, there are mechanisms to provide warnings and advice regarding detected suspicious links and information. These mechanisms provide visual or auditory notifications to older adults and offer specific advice to protect themselves from fraud. Specifically, they may instruct older adults not to click on suspicious links or advise them to avoid accessing suspicious websites. These notifications and advice are provided in a way that is easy for older adults to understand and help them take specific actions to protect themselves from fraud.

[0477] The following describes the processing flow for each example of the form.

[0478] "Example of form 1"

[0479] Step 1: Input the content of suspicious phone calls or emails received by the elderly person into a generative AI.

[0480] Step 2: The generative AI uses natural language processing techniques to analyze the content of phone calls and emails. This analysis includes the analysis of keywords, phrases, and context.

[0481] Step 3: The generative AI determines the likelihood of fraud based on the analysis results. If it determines that there is a high probability of fraud, it warns the elderly about the information.

[0482] "Example of form 2"

[0483] Step 1: Monitor the actions taken by elderly individuals when they use the internet.

[0484] Step 2: Detect any attempts to provide suspicious links or information.

[0485] Step 3: Issue a warning to the elderly person regarding the detected operation.

[0486] "Example of form 3"

[0487] Step 1: If suspicious links or information are detected, provide visual or auditory notifications to the elderly.

[0488] Step 2: The notification will include specific advice on how to protect yourself from fraud.

[0489] Step 3: Provide notifications and advice in a way that is easy for seniors to understand, and help them take concrete actions to protect themselves from fraud.

[0490] (Example 1)

[0491] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0492] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to provide appropriate warnings. However, conventional methods have insufficient information processing capabilities for determining the potential for fraud, resulting in a high risk of elderly people becoming victims of fraud.

[0493] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0494] In this invention, the server includes information processing means, decision means, and identification means. This makes it possible to analyze suspicious communications received by the elderly, assess the possibility of fraud, and provide appropriate warnings.

[0495] "Information processing means" refers to a device or system that has the function of analyzing suspicious communications received by elderly people and extracting important keywords using natural language processing technology.

[0496] A "decision-making tool" is a device or system that has the function of evaluating the possibility of fraud based on the analyzed communication content and generating a warning message based on the result.

[0497] "Identification means" refers to a device or system that has the function of detecting suspicious information provided by an elderly person and identifying whether that information may be related to fraud.

[0498] A "notification means" is a device or system that has the function of issuing a warning about detected suspicious information and providing warning messages to elderly people in a visual or auditory form.

[0499] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and process human language, and is used to extract meaning from text data.

[0500] A "warning message" is a message generated to alert older adults to potentially fraudulent communications, and is provided visually or audibly.

[0501] This invention is a system for analyzing suspicious communications received by elderly people and evaluating the possibility of fraud. The system includes information processing means, judgment means, identification means, and notification means.

[0502] 1. Program generation

[0503] The user generates a generative AI program using natural language processing technology. This program has the function of analyzing suspicious communications received by elderly people and determining the possibility of fraud.

[0504] 2. Program Processing

[0505] The server uses Python's natural language processing libraries, NLTK and spaCy, to analyze the communication content. The server normalizes the text data and extracts important keywords and phrases. Furthermore, it uses a generative AI model to analyze the context and assess the likelihood of fraud.

[0506] 3. Specific examples

[0507] For example, if an elderly person receives an email saying, "Your bank account has been used fraudulently. Please contact us immediately," the server will detect keywords such as "bank account," "fraudulent use," and "urgent," and determine that it is highly likely to be a scam. Based on this information, the server will generate a warning message and notify the user via their device that "This email may be a scam. Please be careful."

[0508] 4. Example of a prompt statement

[0509] An example of a prompt message for a generative AI model is, "Please analyze the content of this email and assess its likelihood of being a scam."

[0510] In this way, the system can respond quickly and accurately to suspicious communications received by the elderly, thereby preventing them from becoming victims of fraud.

[0511] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0512] Step 1:

[0513] The user inputs the content of suspicious communications received by the elderly person into the terminal. The terminal then sends this input data to the server. The input data is in text format.

[0514] Step 2:

[0515] The server preprocesses the received text data. Specifically, it uses Python's natural language processing libraries, NLTK and spaCy, to normalize the text. This includes removing unnecessary symbols and spaces, and converting words to lowercase. As a result of the preprocessing, text data in a format suitable for analysis is output.

[0516] Step 3:

[0517] The server extracts important keywords and phrases from pre-processed text data. This includes using natural language processing techniques to identify words that may be related to fraud. For example, keywords such as "bank account," "fraudulent use," and "urgent" are extracted. The extracted keywords become input data for evaluating the likelihood of fraud.

[0518] Step 4:

[0519] The server uses a generative AI model to analyze the context based on extracted keywords. The AI ​​model understands the meaning of the entire text and assesses the likelihood of fraud. This assessment determines whether the text is likely to be fraudulent. The evaluation results become input data for generating a warning message.

[0520] Step 5:

[0521] The server generates a warning message based on the evaluation results. If it is determined that there is a high probability of fraud, a warning message is created. For example, a message such as "This email may be a scam. Please be careful." is generated. The generated message is sent to the terminal.

[0522] Step 6:

[0523] The device displays warning messages received from the server to the user. Users can receive warnings visually or audibly. This helps older adults to be alerted to potentially fraudulent communications.

[0524] (Application Example 1)

[0525] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0526] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to issue appropriate warnings. However, conventional methods can lead to delays or incorrect judgments, putting the elderly at risk of becoming victims. To solve this problem, technology is needed that can analyze communication content in real time and determine the potential for fraud with high accuracy.

[0527] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0528] In this invention, the server includes information processing means, evaluation means, and identification means. This makes it possible to analyze received communications in real time and quickly display a warning if there is a possibility of fraud.

[0529] "Information processing means" refers to a device or program that analyzes the content of received communications and performs data processing to determine the possibility of fraud.

[0530] An "evaluation tool" is a device or program that scores the likelihood of fraud based on the analyzed communication content and makes a judgment based on the result.

[0531] "Identification means" refers to a device or program that detects suspicious elements from information provided by elderly individuals and issues a warning based on that detection.

[0532] "Notification means" refers to a device or program that provides a visual or auditory warning to a user in the event of potential fraud.

[0533] "Means of real-time analysis" refers to a device or program that instantly analyzes the content of a communication as soon as it is received and quickly outputs the results.

[0534] The system for implementing this invention involves a server and a terminal working in cooperation. The server is equipped with information processing means, evaluation means, and identification means, and analyzes the communication content received from the terminal. Specifically, the server uses a natural language processing library (e.g., spaCy, NLTK) to tokenize the communication content and extract keywords and phrases. Next, it uses a generative AI model (e.g., OpenAI GPT) to analyze the context of the extracted keywords and phrases and score the likelihood of fraud.

[0535] The device receives analysis results from the server and, if it detects a high probability of fraud, provides the user with a visual or auditory warning. This allows the user to respond quickly to suspicious communications.

[0536] As a concrete example, if a user receives an email with the subject line "Important Notice from the Bank," the device sends its contents to the server. The server extracts keywords such as "bank," "important," and "notice," and analyzes them using a generative AI model. If it is determined that the email is highly likely to be fraudulent, the device displays a warning message saying, "This email may be fraudulent. Do not click on the link."

[0537] An example of a prompt message is: "Please analyze the following email content to determine if it is a scam. Email content: 'Important notice from your bank. Please update your account information.'"

[0538] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0539] Step 1:

[0540] The terminal retrieves the content of communications (phone calls and emails) received by the user. This content is sent to the server as text data. The input is the content of communications received by the user, and the output is the transmission of text data to the server.

[0541] Step 2:

[0542] The server tokenizes the received text data using a natural language processing library (e.g., spaCy, NLTK). This process extracts keywords and phrases from the text data. The input is the text data sent from the terminal, and the output is the extracted keywords and phrases.

[0543] Step 3:

[0544] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the context of the extracted keywords and phrases. This analysis scores the likelihood of fraud. The input is the extracted keywords and phrases, and the output is a score indicating the likelihood of fraud.

[0545] Step 4:

[0546] The server determines whether there is a high probability of fraud based on the scoring results. If it determines that there is a high probability of fraud, it generates a warning message. The input is the score indicating the likelihood of fraud, and the output is the warning message.

[0547] Step 5:

[0548] The terminal notifies the user of warning messages received from the server. These notifications are visual or auditory, warning the user of potential fraud. The input is the warning message from the server, and the output is the warning notification to the user.

[0549] (Example 2)

[0550] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0551] When elderly people use the internet, there is a risk of accessing suspicious links or information. Such risks can lead to phishing scams and malware infections, posing a threat to the safety of the elderly. Traditional methods are insufficient to prevent these risks, making it difficult for the elderly to use the internet with peace of mind.

[0552] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0553] In this invention, the server includes information processing means, monitoring means, and notification means. This makes it possible to immediately issue a warning when an elderly person attempts to access suspicious links or information, thereby preventing harm.

[0554] "Information processing means" refers to technology for analyzing suspicious communications received by elderly people and understanding their content.

[0555] "Monitoring measures" refer to technologies that monitor the actions of elderly people in real time and detect suspicious links or information.

[0556] "Notification means" refers to technology that generates warnings for detected suspicious links or information and informs elderly people.

[0557] A "monitoring device" is a device used to monitor the actions of elderly people in real time and has the function of recording the user's behavior.

[0558] "Analysis methods" refer to technologies used to compare suspicious links and information with a database and evaluate their safety.

[0559] "Display means" refers to technologies for displaying warning messages to the elderly visually or audibly.

[0560] This invention is a system for reducing the risk of elderly people accessing suspicious links or information when using the internet. The system includes information processing means, monitoring means, and notification means.

[0561] The server uses information processing tools to analyze suspicious communications received by elderly individuals. This analysis utilizes natural language processing technology to understand the content of the communications. Specifically, the server uses software such as a "natural language processing engine" to analyze the content of emails and messages.

[0562] The terminal uses monitoring devices to monitor the elderly person's actions in real time. Monitoring software installed on the terminal captures links the user attempts to click and URLs of websites they attempt to access, and sends them to a server. Security software is used for this monitoring.

[0563] The server uses analysis tools to compare URLs sent from the terminal against a database of known suspicious links and phishing sites. For example, it uses a "security evaluation API" to assess the security of a URL.

[0564] If suspicious links or information are detected, the server generates a warning message using notification mechanisms. This warning message is sent to the terminal and notified to the elderly visually or audibly through display mechanisms. For example, a message such as "This link is suspicious. Please stop accessing it." might be displayed.

[0565] As a concrete example, consider a scenario where an elderly person opens an email with the subject line "Receive a free gift" and attempts to click on a link within it. The device captures the link and sends it to the server. The server analyzes the link, and if it determines it may be a phishing attempt, it generates a warning message saying, "This link is suspicious. Do not access it," and sends it to the device. The device displays this warning as a pop-up to the user, allowing the user to refrain from clicking the link.

[0566] An example of a prompt to input into the generation AI model would be something like, "Determine whether the link an elderly person is about to click is safe, and generate a warning message if necessary."

[0567] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0568] Step 1:

[0569] Users use the internet.

[0570] Users open emails or browse websites. For example, a user might try to click a link in an email. This action constitutes input.

[0571] Step 2:

[0572] The device monitors the user's actions.

[0573] The terminal uses monitoring software to monitor user actions in real time. Specifically, it captures the URLs of links the user attempts to click and websites they attempt to access. These URLs are then sent to the server as output.

[0574] Step 3:

[0575] The server detects suspicious links and information.

[0576] The server receives the URL sent from the terminal and uses analysis tools to compare it against a database of known suspicious links and phishing sites. This comparison with the database is data processing, and the server evaluates the safety of the URL. The result of this evaluation is the output.

[0577] Step 4:

[0578] The server generates a warning message.

[0579] Based on the analysis results, the server generates a warning message if it detects suspicious links or information. For example, it might create a message such as, "This link is suspicious. Please stop accessing it." This message is then sent to the terminal as output.

[0580] Step 5:

[0581] The device displays a warning to the user.

[0582] The device receives warning messages sent from the server and displays them to the user. Specifically, these are displayed as pop-up notifications or on-screen alerts. This display is the output, alerting the user to the warning. The user can then review the warning and choose a safe course of action.

[0583] (Application Example 2)

[0584] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0585] When elderly people use the internet, there is an increasing risk of fraud and security risks due to accessing suspicious links or information. In this situation, there is a need to provide an environment where elderly people can use the internet with peace of mind.

[0586] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0587] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for issuing warnings and advice regarding the detected suspicious information, a means for scanning links in real time in cooperation with a database for evaluating the safety of links, and a means for displaying warnings based on the evaluation results. This prevents elderly people from accessing suspicious links and information, enabling them to use the internet safely.

[0588] "Elderly people" refers to individuals whose physical or cognitive abilities are likely to decline due to aging.

[0589] "Suspicious communication" refers to means of information transmission, such as phone calls or emails, that contain content that the recipient finds questionable.

[0590] "Generative artificial intelligence methods" refer to technologies that use machine learning and natural language processing to analyze data and make judgments and predictions according to specific purposes.

[0591] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[0592] "Means for detecting suspicious information" refers to technologies for identifying and distinguishing suspicious elements from received information.

[0593] "Means of providing warnings and advice" refers to notification methods that provide users with warnings and guidance on how to act.

[0594] A "database for evaluating link safety" refers to a collection of information used to assess the reliability and safety of links on the internet.

[0595] "Methods for scanning links in real time" refer to technologies that instantly assess the security of a link the moment it is clicked.

[0596] "Means of displaying warnings based on evaluation results" refers to methods for displaying appropriate warnings to users in accordance with the results of link safety evaluations.

[0597] The system for implementing this invention mainly consists of a server and a user terminal. The server uses generative artificial intelligence to analyze suspicious communications received by elderly people and determine if they are fraudulent. Specifically, the server utilizes natural language processing technology to analyze the content of received emails and phone calls. This allows it to identify potentially fraudulent information and issue warnings to the user.

[0598] The user's device has an application installed to evaluate the security of links. This application scans links in real time as they are clicked and evaluates their security in conjunction with a database. Based on the evaluation results, the device displays a visual or audible warning to the user. This helps prevent elderly people from accessing suspicious links.

[0599] For example, if a user attempts to click a link in an email, and the system determines that the link may lead to a phishing site, it will display a warning message such as, "This link is not secure. Please stop accessing it." An example of a prompt message generated using an AI model would be, "Is this link secure? Please evaluate the link's URL."

[0600] In this way, the server and the user's terminal work together to provide an environment that allows elderly people to use the internet safely.

[0601] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0602] Step 1:

[0603] The user receives an email. The email content is displayed on the device. The user attempts to click a link in the email.

[0604] Step 2:

[0605] The device retrieves the URL of the clicked link. It then sends the retrieved URL to the server.

[0606] Step 3:

[0607] The server analyzes the received URL using a generation AI model. Specifically, it uses natural language processing technology to evaluate the structure and content of the URL and determine its potential for fraud. As a result of the analysis, it outputs an evaluation of whether the URL is safe or not.

[0608] Step 4:

[0609] The server sends the URL evaluation result to the device. The evaluation result includes information on whether the link is safe or suspicious.

[0610] Step 5:

[0611] The device displays a warning to the user based on the evaluation results received from the server. Specifically, if a link is deemed suspicious, it displays a warning message such as, "This link is not secure. Please stop accessing it."

[0612] Step 6:

[0613] The user reviews the warning from their device and decides whether to cancel access to the link. If the user cancels access, the device blocks the connection to the link.

[0614] This series of processes prevents elderly people from accessing suspicious links, enabling them to use the internet safely.

[0615] (Example 3)

[0616] Next, we will describe Embodiment 3 of Embodiment Example 3. 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".

[0617] There is a need to provide effective warnings and advice to prevent elderly people from becoming victims of fraud by addressing suspicious communications and links they receive. However, conventional methods are insufficient in detecting suspicious links and providing warnings, and the risk of elderly people becoming victims of fraud still remains. To solve this problem, notifications in a format that elderly people can easily understand and more accurate detection of suspicious links are necessary.

[0618] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0619] In this invention, the server includes generative artificial intelligence means for analyzing suspicious communications received by elderly people, means for determining the possibility of fraud, and means for analyzing suspicious links and evaluating their safety using a machine learning model. This makes it possible to provide elderly people with visually or audibly effective warnings and advice, thereby preventing them from becoming victims of fraud.

[0620] "Generative artificial intelligence means" refers to artificial intelligence technology used to analyze suspicious communications received by elderly people and determine the possibility of fraud.

[0621] "Means of determining the possibility of fraud" refers to the process of analyzing the content of received communications or links and evaluating whether they are related to fraudulent activity.

[0622] "Means for detecting suspicious links and information" refers to technologies that identify suspicious elements within communications and links and issue warnings to users.

[0623] "Means of providing warnings and advice" refers to methods of providing users with messages or guidelines to alert them to suspicious links or information that have been detected.

[0624] "Methods for evaluating safety using machine learning models" refers to the process of analyzing links and information using algorithms that have learned from past data in order to evaluate their safety.

[0625] "Means of providing visual or auditory notification" refers to methods of communicating warnings about suspicious links or information to users through visual displays or audio.

[0626] As an embodiment of this invention, the following system is constructed.

[0627] The server uses a generative AI model to analyze suspicious communications received by elderly individuals. This involves using natural language processing libraries and machine learning frameworks. Specifically, it utilizes software such as "spaCy" and "TensorFlow" to analyze communication content and detect suspicious links and information. The server then uses a machine learning model to evaluate the safety of the detected suspicious links and determine their potential for fraud.

[0628] The device receives warning messages and advice sent from the server and notifies the elderly visually or audibly. Specifically, it uses pop-up messages and audio alerts to warn users not to click on suspicious links.

[0629] The user can initiate the system's operation by entering prompts for the generated AI model. For example, the following prompts might be used:

[0630] "Please create a program that can determine if links in emails received by elderly people are safe and display a warning if necessary."

[0631] Based on this prompt, the generating AI model creates a program to detect suspicious links and issue warnings. This enables elderly people to take concrete actions to prevent themselves from becoming victims of fraud. The flow of the specific processing in Example 3 will be explained using Figure 15.

[0632] Step 1:

[0633] The user inputs a prompt into the generative AI model. For example, the prompt might be, "Create a program that determines whether a link in an email received by an elderly person is safe and displays a warning if necessary." Based on this input, the generative AI model generates a program to detect suspicious links.

[0634] Step 2:

[0635] The server uses the generated program to analyze emails and communications received by elderly individuals. The input includes email content and link information. The server uses the natural language processing library "spaCy" to analyze the email content and extract links and domain information. The output provides potential suspicious links.

[0636] Step 3:

[0637] The server evaluates the extracted link information using a machine learning model. The input includes the suspicious link candidates obtained in step 2. The server uses TensorFlow to assess the safety of the links and determine their likelihood of being fraudulent. The output is a list of suspicious links.

[0638] Step 4:

[0639] The server generates warning messages and advice for links it deems suspicious. The input includes the list of suspicious links obtained in step 3. The server creates a warning message for the user, such as "This link is suspicious. Do not click it." The output is the generated warning message.

[0640] Step 5:

[0641] The server sends the generated warning message to the terminal. The input includes the warning message generated in step 4. The server prepares to send the warning message to the terminal and notify the user. The output is that the message has been sent to the terminal.

[0642] Step 6:

[0643] The terminal displays warning messages received from the server to the user. Input includes the warning messages sent from the server. The terminal provides visual or auditory notifications to the elderly using pop-up messages or audio alerts. Output is the completion of the warning to the user.

[0644] (Application Example 3)

[0645] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0646] One challenge when elderly people use digital communications is preventing them from becoming victims of fraud or malicious links. In particular, elderly people have difficulty distinguishing suspicious links and information, which increases their risk of becoming victims of fraud. There is a need to improve this situation and provide an environment where elderly people can use digital communications with peace of mind.

[0647] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0648] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for scanning communication links in real time and issuing a warning when a suspicious link is detected, and a means for issuing warnings through visual notifications or audio alerts. This makes it possible for elderly people to easily identify suspicious links and information and prevent themselves from becoming victims of fraud.

[0649] "Elderly people" generally refers to older individuals who require support in using digital technologies.

[0650] "Suspicious communications" refers to means of information transmission such as phone calls, emails, and messages that may be involved in fraud or illegal activities.

[0651] "Generative artificial intelligence" refers to artificial intelligence technologies used to analyze data and detect specific patterns or anomalies.

[0652] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[0653] "Means for detecting suspicious information" refers to technologies for identifying fraudulent or questionable elements from received information.

[0654] "Real-time scanning" refers to a technology that analyzes information instantly upon receipt and provides the results.

[0655] "Warning mechanisms" refer to technologies that provide notifications and alerts to inform users of potential dangers.

[0656] "Visual notifications" refer to means of conveying information through visual means, such as messages and pop-ups displayed on the screen.

[0657] "Voice alert" refers to a warning method that uses sound to draw the user's attention.

[0658] The system for implementing this invention mainly consists of a server and a terminal. The server uses generative artificial intelligence to analyze suspicious communications received by elderly people and determine if they are fraudulent. Specifically, the server utilizes natural language processing technology to analyze the content of communications. This allows it to detect suspicious information and perform real-time scans.

[0659] The device scans the links of communications received by the user in real time and issues a warning if it detects a suspicious link. The warning is provided to the user through visual notifications and audio alerts. Visual notifications appear as pop-ups on the device screen, and audio alerts are generated using a Text-to-Speech (TTS) engine.

[0660] As a concrete example, a device automatically scans a link received by a user via SMS and displays a warning message saying, "This link may lead to a phishing site. Do not click it." In this process, the server uses the Google Safe Browsing API to evaluate the safety of the link.

[0661] An example of a prompt to input into a generating AI model is, "Is this link safe? If it may be a phishing site, please display a warning." Using this prompt, the system can provide appropriate warnings to the user.

[0662] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[0663] Step 1:

[0664] The user receives the communication on their device. The device extracts links and text data to analyze the content of the received communication. The input is the received communication data, and the output is the links and text data to be analyzed.

[0665] Step 2:

[0666] The device sends extracted links to the server for real-time scanning. The server uses the Google Safe Browsing API to evaluate the safety of the links. The input is the links to be analyzed, and the output is the evaluation result regarding the safety of the links.

[0667] Step 3:

[0668] The server uses generative artificial intelligence to analyze the text data of received communications using natural language processing. This allows it to determine whether the communications are potentially fraudulent. The input is the text data of the communications, and the output is the evaluation result regarding the likelihood of fraud.

[0669] Step 4:

[0670] The server integrates the link safety assessment results and the text data analysis results, and generates a warning if suspicious information is detected. The input is the link safety assessment results and the fraud potential assessment results, and the output is a warning message.

[0671] Step 5:

[0672] The terminal notifies the user of warning messages received from the server. It displays a pop-up on the screen as a visual notification and plays the warning as an audio alert using a Text-to-Speech (TTS) engine. The input is the warning message, and the output is the warning notification to the user.

[0673] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0674] "Example of form 1"

[0675] One embodiment of the present invention involves a system that combines a generative AI for analyzing suspicious phone calls and emails received by elderly people, means for determining the possibility of fraud, means for detecting suspicious links and information provided by elderly people, and means for providing warnings and advice regarding detected suspicious links and information, with an emotion engine. This emotion engine analyzes the emotions of elderly people from their tone of voice and facial expressions, and incorporates the results as part of the fraud judgment. Specifically, it analyzes the tone and speed of the elderly person's voice when they speak on the phone, or changes in facial expressions read from video calls or photographs, and reflects the results in the fraud judgment.

[0676] "Example of form 2"

[0677] Furthermore, the emotional engine adjusts how warnings and advice are delivered based on the emotions of the elderly. For example, if it determines that an elderly person is feeling anger or confusion, it uses that information to adjust the tone and expression of warnings and advice, providing information in a way that is easy for the elderly person to understand. This makes it possible to provide appropriate advice and protect elderly people from fraud even when they are emotionally charged.

[0678] "Example of form 3"

[0679] Furthermore, the emotion engine can track changes in the emotions of older adults over the long term and incorporate that information into fraud detection. For example, if an older adult is consistently judged to be experiencing anxiety or fear, this information can be used to increase the likelihood of fraud and encourage more careful monitoring. This allows for the use of changes in the emotions of older adults in fraud detection, resulting in more accurate fraud detection.

[0680] The following describes the processing flow for each example of the form.

[0681] "Example of form 1"

[0682] Step 1: A generative AI analyzes suspicious phone calls and emails received by elderly individuals.

[0683] Step 2: Based on the analysis results, determine the likelihood of fraud.

[0684] Step 3: Detect suspicious links and information provided by elderly individuals.

[0685] Step 4: The emotion engine analyzes the emotions of the elderly person from their tone of voice and facial expressions.

[0686] Step 5: Use the analysis results as part of the fraud determination.

[0687] "Example of form 2"

[0688] Step 1: The emotion engine analyzes the emotions of the elderly.

[0689] Step 2: Based on the analysis results, adjust how warnings and advice are provided.

[0690] Step 3: Provide adjusted warnings and advice to the elderly.

[0691] "Example of form 3"

[0692] Step 1: The emotion engine tracks changes in older adults' emotions over the long term.

[0693] Step 2: Use the tracking results to determine if it's a scam.

[0694] Step 3: Based on the results, increase the likelihood of fraud and monitor more carefully.

[0695] (Example 1)

[0696] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0697] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to provide appropriate warnings and advice. However, conventional technologies are insufficient in identifying fraud or may delay warnings, which presents a challenge as the elderly are at a high risk of becoming victims of fraud.

[0698] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0699] In this invention, the server includes information processing means, evaluation means, identification means, sentiment analysis means, and notification means. This makes it possible to quickly analyze suspicious communications received by the elderly, determine the possibility of fraud with high accuracy, and provide appropriate warnings and advice.

[0700] "Information processing means" refers to technical means for analyzing suspicious communications received by elderly people, and has the function of analyzing the content of communications using natural language processing technology.

[0701] "Evaluation means" refers to technical means for determining the possibility of fraud based on the analyzed communication content, and has the function of scoring the possibility of fraud using a generative AI model.

[0702] "Identification means" refers to technical means for detecting suspicious information provided by elderly individuals, and includes the function of evaluating the security of links and information.

[0703] "Emotional analysis means" refers to technical means for analyzing the emotions of elderly people and reflecting the results in fraud detection. These means have the function of analyzing emotions using voice recognition technology and image analysis technology.

[0704] "Notification means" refers to technical means for issuing warnings or advice regarding detected suspicious information, and has the function of providing visual or auditory notifications to the elderly.

[0705] This invention is a system for analyzing suspicious communications received by elderly people and determining the possibility of fraud. The system includes information processing means, evaluation means, identification means, sentiment analysis means, and notification means.

[0706] The server uses software such as Python and TensorFlow as information processing tools to analyze the content of suspicious communications received by elderly individuals using natural language processing techniques. This makes it possible to extract keywords and phrases contained in the communications and analyze the context.

[0707] As an evaluation method, the server uses a generative AI model to score the likelihood of fraud based on the analyzed data. If it is determined that there is a high probability of fraud, the process proceeds to the next step.

[0708] The identification mechanism works to detect suspicious links and information provided by users. The server evaluates the safety of the link and issues a warning if it determines that it is dangerous.

[0709] As a means of emotion analysis, the server uses speech recognition and image analysis technologies to analyze the tone and speed of the elderly person's voice, as well as changes in their facial expressions. This allows the server to determine whether the elderly person is feeling anxious or tense, and this information is then used to determine if they are a fraud suspect.

[0710] As a notification method, the server sends warnings and advice to the terminal. The terminal provides elderly users with visual or auditory notifications and displays specific advice. For example, it may display a message such as, "This call may be a scam. Do not provide any personal information."

[0711] For example, if a user receives a suspicious email, their device forwards the email to the server. The server analyzes the email's content, and if it detects phrases such as "verify bank account information," it warns the user, "This email may be a scam. Do not click the link."

[0712] Examples of prompt messages include the following:

[0713] "Could this phone call be a scam?"

[0714] "Are the links in the email safe?"

[0715] "Does this tone of voice indicate anxiety?"

[0716] In this way, the system analyzes suspicious information received by the elderly, determines the possibility of fraud, and issues a warning if necessary.

[0717] The flow of the specific processing in Example 1 will be explained using Figure 17.

[0718] Step 1:

[0719] When a user receives a suspicious phone call or email, they input the content into their device. The device then sends the audio or text data to the server. Specifically, data is collected using the smartphone's recording function or email app. The input is audio files or text data, and the output is data transfer to the server.

[0720] Step 2:

[0721] The server analyzes the received data using natural language processing techniques. Using Python and TensorFlow, it extracts keywords and phrases from phone calls and emails and analyzes the context. Input is audio files or text data, and output is a list of the analyzed keywords and phrases.

[0722] Step 3:

[0723] The server determines the likelihood of fraud based on the analysis results. Using a generative AI model, it scores the likelihood of fraud from detected keyword and phrase combinations. The input is a list of analyzed keywords and phrases, and the output is a score indicating the likelihood of fraud.

[0724] Step 4:

[0725] The server analyzes the user's emotions using an emotion engine. It uses speech recognition technology to analyze the tone and speed of the user's voice, and image analysis technology to read changes in facial expressions from video calls and photos. The input is audio data and image data, and the output is the result of the emotion analysis.

[0726] Step 5:

[0727] The server will issue a warning to the user if it determines that the call is likely to be a scam. It will send a notification to the device, displaying specific advice and warnings. The input is a score indicating the likelihood of fraud and the result of sentiment analysis, and the output is a warning message to the user. Specifically, it will display a message such as, "This call may be a scam. Do not provide any personal information."

[0728] (Application Example 1)

[0729] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0730] There is a need to quickly and accurately determine the possibility of fraud in suspicious communications received by the elderly and to provide appropriate warnings and advice. However, conventional technologies have limitations in analyzing the content of communications and do not take into account the emotional state of the elderly when making fraud judgments, making it difficult to prevent fraud.

[0731] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0732] In this invention, the server includes a generative information processing means for analyzing suspicious communications received by an elderly person, an information processing means for determining the possibility of fraud, and an information processing means for analyzing the elderly person's voice and facial expressions to detect emotions. This makes it possible to make fraud judgments that take into account the emotional state of the elderly person.

[0733] "Elderly people" refers to individuals whose physical or cognitive abilities are likely to decline as a result of aging.

[0734] "Suspicious communication" refers to means of transmitting information, such as phone calls or emails, that may have questionable intentions or content directed at the recipient.

[0735] "Generative information processing means" refers to information processing technology that analyzes received information and makes decisions or generates information according to a specific purpose.

[0736] "Information processing means for determining the possibility of fraud" refers to information processing technology that analyzes received information and evaluates whether that information is related to fraudulent activity.

[0737] "Information processing means for detecting suspicious information" refers to information processing technology for identifying and detecting suspicious elements from received information.

[0738] "Information processing means for issuing warnings and advice" refers to information processing technology that provides recipients with warnings and guidance on action based on detected suspicious information.

[0739] "Information processing means for analyzing voice and facial expressions to detect emotions" refers to information processing technology that analyzes the voice and facial expressions of a recipient to identify their emotional state.

[0740] "Information processing means that reflect the results of emotional analysis in fraud judgment" refers to information processing technology that incorporates analyzed emotional information into the process of evaluating the likelihood of fraud.

[0741] The system for implementing this invention mainly consists of a server and a terminal. The server is equipped with generative information processing means for analyzing suspicious communications received by elderly people, and analyzes the content of the communications using natural language processing technology. Specifically, it uses natural language processing libraries (e.g., spaCy or NLTK) to extract keywords and phrases from the communications and evaluate the possibility of fraud.

[0742] Furthermore, the server is equipped with information processing capabilities to analyze the voice and facial expressions of elderly individuals and detect their emotions. This involves using emotion analysis libraries (such as OpenCV or DeepFace) to analyze audio and video data acquired through the device's camera and microphone. The analysis results are used to determine if a fraud is likely, and if there is a high probability of fraud, warnings and advice are displayed on the device.

[0743] The device sends communications received by the user to a server and receives analysis results from the server. The user receives visual or auditory notifications on the device to be alerted to potentially fraudulent communications.

[0744] For example, if a user receives an email titled "Important Notice from the Bank," the device sends its contents to a server. The server uses natural language processing to analyze keywords such as "bank," "important," and "notice" to determine if it is a scam. Additionally, when a user receives a phone call, the device's camera analyzes their facial expressions to detect changes in emotion. If it is determined that the email is likely to be a scam, a warning message such as "This email may be a scam. Do not click the link." is displayed on the device.

[0745] An example of a prompt to input into the generating AI model is: "Analyze the contents of this email and determine if it is a scam. The email contains the following: 'Important notice from the bank.'"

[0746] The flow of a specific process in Application Example 1 will be explained using Figure 18.

[0747] Step 1:

[0748] The terminal sends suspicious communications (phone calls and emails) received by the user to the server. The input is the content of the communication, and the output is the data to be sent to the server. The terminal converts the communication content into text format and generates data packets for transmission to the server.

[0749] Step 2:

[0750] The server analyzes the received communication content using generative information processing means. The input is the communication content sent from the terminal, and the output is the analysis result. The server uses natural language processing libraries (e.g., spaCy or NLTK) to extract keywords and phrases from the communication content and evaluate the likelihood of fraud.

[0751] Step 3:

[0752] The server analyzes the user's voice and facial expressions to detect emotions. The input is audio and video data sent from the terminal, and the output is the emotion analysis result. The server uses emotion analysis libraries (e.g., OpenCV or DeepFace) to analyze changes in voice tone and facial expressions.

[0753] Step 4:

[0754] The server integrates the results of communication content analysis and sentiment analysis to determine the likelihood of fraud. The input is the results of communication content analysis and sentiment analysis, and the output is the evaluation of the likelihood of fraud. The server uses a generative AI model to make a comprehensive judgment on the likelihood of fraud.

[0755] Step 5:

[0756] The server sends a warning or advice to the terminal if it determines that there is a high probability of fraud. The input is the result of the fraud probability assessment, and the output is the warning message. The server generates the warning message and creates a data packet to send to the terminal.

[0757] Step 6:

[0758] The terminal notifies the user of warnings and advice received from the server. Input is the warning message from the server, and output is the notification to the user. The terminal displays or plays the warning message to the user visually or audibly.

[0759] (Example 2)

[0760] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0761] When elderly people use the internet, there is a problem in that they are at a higher risk of becoming victims of fraud because they cannot respond appropriately to suspicious communications and information. In addition, they may not be able to correctly understand information because they are not provided with appropriate warnings and advice that are tailored to their emotional state.

[0762] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0763] In this invention, the server includes a generative AI means for analyzing suspicious communications received by elderly people, means for determining the possibility of fraud, and means for detecting suspicious information provided by elderly people. This enables elderly people to respond quickly and appropriately to suspicious information and prevent them from becoming victims of fraud.

[0764] "Generative AI methods" refer to artificial intelligence technologies used to analyze suspicious communications received by elderly people and to understand their content.

[0765] "Means for determining the possibility of fraud" refers to algorithms or processes that evaluate whether a communication is fraudulent based on its analyzed content.

[0766] "Methods for detecting suspicious information" refer to technologies for identifying information that exhibits unusual patterns or has known risks from information provided by elderly individuals.

[0767] "Warning measures" refer to methods of providing visual or auditory notifications to alert older adults to detected suspicious information.

[0768] "Methods for analyzing emotional states" refer to techniques for analyzing the tone of voice and facial expressions of elderly people to determine their emotional state.

[0769] "Means of adjusting the content of warnings" refers to methods of modifying the tone and expression of warnings and advice to make them easier for older adults to understand, based on their analyzed emotional state.

[0770] This invention is a system that enables elderly people to appropriately respond to suspicious communications and information when using the internet. The system consists of three components: a server, a terminal, and a user.

[0771] The server uses generative AI methods to analyze suspicious communications received by elderly individuals. This analysis employs natural language processing techniques to understand the content of the communications and determine if they are fraudulent. The server runs algorithms that cross-reference information with a database to identify information that poses a known risk.

[0772] The device uses voice recognition software and a facial recognition camera to analyze the emotional state of elderly individuals. This allows the device to analyze the tone of voice and facial expressions of elderly individuals in real time to determine their emotional state. Based on this emotional state, the device adjusts the content of warnings and advice, providing information in a way that is easy for the elderly to understand.

[0773] Elderly users receive warnings and advice from the device and take appropriate action. For example, when an elderly person inputs the prompt "Is it safe to click this link?" into the AI ​​model, the system evaluates the safety of the link and returns appropriate advice. This allows elderly people to use the internet with peace of mind.

[0774] The flow of the specific processing in Example 2 will be explained using Figure 19.

[0775] Step 1:

[0776] The server monitors communications received by elderly individuals and retrieves emails and website links as input. Using generative AI, the server analyzes the content of these communications using natural language processing techniques. This analysis generates data to determine the likelihood of a communication being fraudulent. The output provides the analysis results and a flag indicating whether the communication is potentially fraudulent.

[0777] Step 2:

[0778] The server detects suspicious information based on the analysis results. It uses a database of analyzed communication content and information with known risks as input. The server compares these and runs algorithms to identify suspicious links and information. A list of suspicious information is generated as output.

[0779] Step 3:

[0780] The terminal receives a list of suspicious information sent from the server and uses it as input. The terminal uses an emotion engine to analyze the emotional state of elderly people. Specifically, it uses speech recognition software and a facial recognition camera to analyze the tone of voice and facial expressions of elderly people in real time. The output is an evaluation result of their emotional state.

[0781] Step 4:

[0782] The device adjusts the content of warnings and advice based on the results of an emotional state assessment. It uses a list of suspicious information and the results of the emotional state assessment as input. The device modifies the tone and expression of warnings and advice to make them easily understandable to older adults. The adjusted warnings and advice are generated as output and provided to older adults as visual or auditory notifications.

[0783] Step 5:

[0784] The elderly user receives warnings and advice from the device and takes appropriate action. Specifically, the elderly user can input a prompt message such as "Is it safe to click this link?" into the AI ​​model. Based on this prompt, the system evaluates the safety of the link and provides feedback to the user. As output, the elderly user is presented with advice regarding the safety of the link.

[0785] (Application Example 2)

[0786] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0787] Fraud cases involving elderly people accessing suspicious links and information online are on the rise. In this situation, there is a need to provide an environment where elderly people can use the internet with peace of mind. Furthermore, it is important to provide appropriate warnings and advice that take into account the feelings of the elderly.

[0788] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0789] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for issuing warnings and advice regarding the detected suspicious information, an emotion analysis means for analyzing the emotions of elderly people and adjusting the expression of warnings and advice, and a means for verifying the safety of suspicious information using an external database. This makes it possible to reduce the risk of elderly people accessing suspicious information and to provide appropriate warnings and advice that take their emotions into consideration.

[0790] "Elderly people" refers to individuals who are considered to be prone to a decline in physical or cognitive abilities as a result of aging.

[0791] "Suspicious communication" refers to means of information transmission such as phone calls and emails that may have been sent to the recipient with an improper purpose.

[0792] "Generative artificial intelligence means" refers to methods that use artificial intelligence technology to analyze data and make judgments or predictions according to specific purposes.

[0793] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[0794] "Means of detecting suspicious information" refers to methods for identifying information that is deemed unreliable or dangerous from among the information provided on the internet.

[0795] "Means of providing warnings and advice" refers to means of generating notifications and messages to draw attention to users or provide guidance on how to act.

[0796] "Emotional analysis tools" refer to methods for analyzing a user's emotional state and taking appropriate action based on the results.

[0797] "Methods of verification using external databases" refer to methods of verifying information by referring to external databases in order to evaluate the safety and reliability of information on the internet.

[0798] The system for implementing this invention mainly consists of a server and a user terminal. The server uses generative artificial intelligence means to analyze suspicious communications received by elderly people. Specifically, a program developed using Python utilizes the Google Cloud Natural Language API to analyze the content of the communications using natural language processing. This determines whether the communications are likely to be fraudulent.

[0799] On the user's device, it is implemented as an internet browser extension to detect suspicious information. It uses the VirusTotal API to verify the safety of links and information. For detected suspicious information, a notification is displayed to provide warnings and advice. This notification is delivered in an appropriate tone, based on an analysis of the user's emotional state using sentiment analysis tools. The Google Cloud Natural Language API is used for sentiment analysis.

[0800] For example, if an elderly person attempts to click on a link in an email, the system may detect that the link is suspicious. If the sentiment analysis system determines that the user is confused, a gentle warning message will be displayed, such as, "This link may not be safe. Please check the details."

[0801] An example of a prompt message is, "The user is about to click a link. Please verify the safety of this link and display a warning appropriate to the user's sentiment." This prompt message allows the system to take appropriate action and protect elderly users from suspicious information.

[0802] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[0803] Step 1:

[0804] The user attempts to click a link on their device. The device retrieves the URL of the clicked link. This URL becomes the input data.

[0805] Step 2:

[0806] The device sends the retrieved URL to the VirusTotal API for security verification. The VirusTotal API evaluates the URL's security and outputs the result. The output indicates whether the URL is safe or suspicious.

[0807] Step 3:

[0808] The server receives evaluation results from the VirusTotal API, and if it determines that the URL is suspicious, it collects sentiment data from the user's device to analyze the user's sentiment. This data includes the user's recent activity history and input content.

[0809] Step 4:

[0810] The server sends the collected emotion data to the Google Cloud Natural Language API for emotion analysis. The input data is information about the user's emotions, and the output is the analysis result showing the user's emotional state.

[0811] Step 5:

[0812] The server generates a warning message in an appropriate tone based on the sentiment analysis results. It uses a generative AI model to create prompts that respond to the user's emotions. For example, if the user is confused, it might generate a message such as, "This link may not be safe. Please check the details."

[0813] Step 6:

[0814] The device displays warning messages received from the server to the user. This allows the user to receive warnings about suspicious links and take appropriate action.

[0815] (Example 3)

[0816] Next, we will describe Embodiment 3 of Embodiment Example 3. 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".

[0817] Reducing the risk of elderly people becoming victims of fraud is crucial. In particular, it is important to appropriately identify suspicious communications and information and issue prompt warnings. Improving the accuracy of fraud detection, taking into account the emotional changes of the elderly, is also a challenge.

[0818] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[0819] In this invention, the server includes information processing means, judgment means, detection means, notification means, and emotion analysis means. This makes it possible to analyze suspicious communications received by elderly people, assess the possibility of fraud, and provide appropriate warnings and advice. Furthermore, by tracking changes in the elderly person's emotions and reflecting them in fraud judgments, the accuracy of fraud detection can be improved.

[0820] "Information processing means" refers to technology for analyzing suspicious communications received by elderly people and understanding their content.

[0821] "Decision-making tools" are techniques used to evaluate the likelihood of fraud based on analyzed information.

[0822] "Detection means" refers to technology for identifying and distinguishing suspicious information provided by elderly individuals.

[0823] "Notification methods" refer to technologies used to communicate warnings and advice to the elderly regarding detected suspicious information.

[0824] "Emotional analysis methods" refer to technologies that track changes in the emotions of elderly people and use that information to make fraud judgments.

[0825] This invention is a fraud detection system for the elderly, realized through the respective roles of the server, terminal, and user. The server uses information processing means, judgment means, detection means, notification means, and sentiment analysis means to analyze suspicious communications received by the elderly and evaluate the possibility of fraud.

[0826] Specifically, the server uses a generative AI model to scan links and information contained in emails and messages to detect suspicious elements. For example, it analyzes the URL of a link and compares it against a list of known fraudulent sites to determine the likelihood of fraud. It also uses an emotion engine to collect user emotion data and track changes in their emotions. This allows the system to incorporate information about when a user is feeling anxious or fearful into its fraud detection.

[0827] The device provides visual or auditory notifications to the elderly based on instructions from the server. For example, if a user attempts to click a suspicious link, the device displays a warning such as, "This link may be a scam. Do not click it." It also provides advice based on emotional data, such as, "You seem to have been feeling anxious recently. Be careful of scams."

[0828] Users receive notifications from their devices and take specific actions. For example, they may be prompted not to click on a link they are warned about or to delete a suspicious email.

[0829] An example of a prompt message is, "Please determine if the link in this email is safe." The generating AI model then analyzes the link's safety and notifies the user of the result. Figure 21 illustrates the specific processing flow in Example 3.

[0830] Step 1:

[0831] Users generate data through everyday actions. Inputs include receiving emails, browsing websites, and sending and receiving messages. This data is sent to a server. The server collects this data and prepares it for analysis. Specifically, it stores email content and link information in a database.

[0832] Step 2:

[0833] The server analyzes the collected data. The input is data from emails and messages sent by users. The server uses a generative AI model to scan links and information contained in emails and messages, detecting suspicious elements. Specifically, it analyzes the URLs of links and compares them against a list of known fraudulent websites. The output is a list of suspicious links and information.

[0834] Step 3:

[0835] The server uses an emotion engine to collect user emotion data. Input includes user behavior and input data while using the device. The server analyzes this data to infer changes in the user's emotions. Specifically, if the user frequently exhibits anxiety-inducing behaviors, this data is recorded. The output is a report on changes in emotions.

[0836] Step 4:

[0837] The server integrates analysis results and sentiment data to determine the likelihood of fraud. Inputs include a list of suspicious links and information, and reports on changes in sentiment. Based on this data, the server assesses the risk of fraud and generates warnings as needed. The output is a warning message.

[0838] Step 5:

[0839] The device provides visual or auditory notifications to the user based on instructions from the server. The input is a warning message from the server. If the user attempts to click a suspicious link, the device will display a warning such as, "This link may be fraudulent. Do not click it." Specifically, this involves displaying a large red warning message on the screen. The output is a warning notification to the user.

[0840] Step 6:

[0841] The user receives a notification from their device and takes specific action. The input is a warning notification from the device. The user is prompted to take action such as not clicking the warned link or deleting a suspicious email. The specific action is to choose an action to avoid the risk of fraud. The output is that the user's safe actions are promoted.

[0842] (Application Example 3)

[0843] Next, we will describe application example 3 of form example 3. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0844] There is a need to reduce the risk of elderly people becoming victims of fraud. However, conventional methods have difficulty in appropriately detecting and warning about suspicious communications and information, and fraud judgments do not take into account the emotional changes of the elderly, resulting in insufficient accuracy in fraud prevention.

[0845] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0846] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, and an emotion analysis means for tracking changes in the elderly person's emotions and reflecting them in the fraud determination. This reduces the risk of fraud against elderly people and enables more accurate fraud prevention.

[0847] "Generative artificial intelligence methods" refer to methods that use artificial intelligence technology to analyze suspicious communications received by elderly people and determine the possibility of fraud.

[0848] "Means of determining the possibility of fraud" refers to methods of evaluating whether the analyzed information is related to fraud.

[0849] "Means for detecting suspicious information" refers to methods for identifying suspicious elements in information provided by elderly individuals and issuing warnings.

[0850] "Means of providing warnings or advice" refers to means of visually or audibly drawing the attention of elderly individuals to any suspicious information that has been detected.

[0851] "Emotional analysis methods" refer to techniques for tracking changes in the emotions of elderly individuals and using that information to inform fraud detection.

[0852] "Notification means" refers to methods for sending visual or auditory warnings to elderly people.

[0853] The system for implementing this invention mainly consists of a server and a terminal. The server is equipped with generative artificial intelligence means for analyzing suspicious communications received by elderly people. This means analyzes the content of communications using natural language processing technology and determines the possibility of fraud. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to build a model for analyzing the content of communications.

[0854] The devices in question are smartphones, tablets, and other devices used by the elderly, and these devices have notification mechanisms implemented. These notification mechanisms issue visual or auditory warnings based on analysis results. For example, visual warnings can be displayed using Android Notifications or iOS Push Notifications, and voice alerts can be provided using Google Assistant or Siri.

[0855] Furthermore, the server is equipped with emotion analysis capabilities to track changes in the emotions of elderly individuals. This capability collects voice and facial expression data from elderly individuals and uses an AI model to analyze changes in their emotions. This makes it possible to incorporate emotional information into fraud detection.

[0856] For example, if an elderly person opens an email on their smartphone and it contains a suspicious link, the device will issue a voice warning saying, "This link may be a scam. Do not click it," and display a red warning message on the screen.

[0857] An example of a prompt message would be, "Is this link safe? If it may be a scam, please generate a warning message." This would enable older adults to take concrete actions to protect themselves from fraud.

[0858] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[0859] Step 1:

[0860] The server receives communication data received by elderly individuals. The input consists of email and phone call content, which is then passed to a generative artificial intelligence system. The server analyzes the communication content using natural language processing techniques to determine the likelihood of fraud. The output is the result of the determination of whether or not fraud is likely.

[0861] Step 2:

[0862] The server detects suspicious information based on the analysis results. The input is the judgment result obtained in step 1, which is used to identify suspicious links and information. The server lists the identified suspicious information and sends it to the terminal. The output is a list of suspicious information.

[0863] Step 3:

[0864] The terminal issues a warning to the user based on a list of suspicious information received from the server. The input is a list of suspicious information, which is used to provide visual or auditory warnings. The terminal displays a warning message on the screen and uses a voice assistant to draw attention. The output is the notification of the warning to the user.

[0865] Step 4:

[0866] The server collects emotional data from elderly individuals and analyzes it using emotion analysis tools. Inputs include voice and facial expression data, which are then analyzed by an AI model. The server tracks emotional changes and incorporates them into fraud detection. The output is the result of the emotion analysis.

[0867] Step 5:

[0868] The server improves the accuracy of its fraud detection based on the sentiment analysis results. The input consists of the sentiment analysis results and the judgment results from Step 1, which are integrated to assess the final fraud risk. The server sends the assessment results to the terminal and strengthens warnings as needed. The output is the final fraud risk assessment.

[0869] (Other examples)

[0870] Since this is the same as the specific processing described in the other embodiments of the first embodiment above, the explanation will be omitted.

[0871] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0872] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0873] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[0874] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0875] [Third Embodiment]

[0876] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0877] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0878] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0879] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0880] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0881] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0882] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0883] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0884] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0885] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0886] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0887] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0888] "Example of form 1"

[0889] One embodiment of the present invention involves a generative AI system that analyzes suspicious phone calls and emails received by elderly individuals. This generative AI uses natural language processing technology to analyze the content of phone calls and emails and determine the likelihood of them being scams. Specifically, it analyzes keywords, phrases, and context contained in the content of phone calls and emails, and if it determines that there is a high probability of fraud, it warns the elderly individual with that information.

[0890] "Example of form 2"

[0891] Furthermore, there are methods to detect suspicious links and information provided by elderly individuals. These methods monitor the actions of elderly individuals when they use the internet, and if they attempt to provide suspicious links or information, they detect the action and warn the elderly individual. Specifically, they detect actions such as elderly individuals attempting to click on links in emails or access suspicious websites, and warn the elderly individual.

[0892] "Example of form 3"

[0893] Furthermore, there are mechanisms to provide warnings and advice regarding detected suspicious links and information. These mechanisms provide visual or auditory notifications to older adults and offer specific advice to protect themselves from fraud. Specifically, they may instruct older adults not to click on suspicious links or advise them to avoid accessing suspicious websites. These notifications and advice are provided in a way that is easy for older adults to understand and help them take specific actions to protect themselves from fraud.

[0894] The following describes the processing flow for each example of the form.

[0895] "Example of form 1"

[0896] Step 1: Input the content of suspicious phone calls or emails received by the elderly person into a generative AI.

[0897] Step 2: The generative AI uses natural language processing techniques to analyze the content of phone calls and emails. This analysis includes the analysis of keywords, phrases, and context.

[0898] Step 3: The generative AI determines the likelihood of fraud based on the analysis results. If it determines that there is a high probability of fraud, it warns the elderly about the information.

[0899] "Example of form 2"

[0900] Step 1: Monitor the actions taken by elderly individuals when they use the internet.

[0901] Step 2: Detect any attempts to provide suspicious links or information.

[0902] Step 3: Issue a warning to the elderly person regarding the detected operation.

[0903] "Example of form 3"

[0904] Step 1: If suspicious links or information are detected, provide visual or auditory notifications to the elderly.

[0905] Step 2: The notification will include specific advice on how to protect yourself from fraud.

[0906] Step 3: Provide notifications and advice in a way that is easy for seniors to understand, and help them take concrete actions to protect themselves from fraud.

[0907] (Example 1)

[0908] Next, we will describe Embodiment 1 of Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0909] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to provide appropriate warnings. However, conventional methods have insufficient information processing capabilities for determining the potential for fraud, resulting in a high risk of elderly people becoming victims of fraud.

[0910] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0911] In this invention, the server includes information processing means, decision means, and identification means. This makes it possible to analyze suspicious communications received by the elderly, assess the possibility of fraud, and provide appropriate warnings.

[0912] "Information processing means" refers to a device or system that has the function of analyzing suspicious communications received by elderly people and extracting important keywords using natural language processing technology.

[0913] A "decision-making tool" is a device or system that has the function of evaluating the possibility of fraud based on the analyzed communication content and generating a warning message based on the result.

[0914] "Identification means" refers to a device or system that has the function of detecting suspicious information provided by an elderly person and identifying whether that information may be related to fraud.

[0915] A "notification means" is a device or system that has the function of issuing a warning about detected suspicious information and providing warning messages to elderly people in a visual or auditory form.

[0916] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and process human language, and is used to extract meaning from text data.

[0917] A "warning message" is a message generated to alert older adults to potentially fraudulent communications, and is provided visually or audibly.

[0918] This invention is a system for analyzing suspicious communications received by elderly people and evaluating the possibility of fraud. The system includes information processing means, judgment means, identification means, and notification means.

[0919] 1. Program generation

[0920] The user generates a generative AI program using natural language processing technology. This program has the function of analyzing suspicious communications received by elderly people and determining the possibility of fraud.

[0921] 2. Program Processing

[0922] The server uses Python's natural language processing libraries, NLTK and spaCy, to analyze the communication content. The server normalizes the text data and extracts important keywords and phrases. Furthermore, it uses a generative AI model to analyze the context and assess the likelihood of fraud.

[0923] 3. Specific examples

[0924] For example, if an elderly person receives an email saying, "Your bank account has been used fraudulently. Please contact us immediately," the server will detect keywords such as "bank account," "fraudulent use," and "urgent," and determine that it is highly likely to be a scam. Based on this information, the server will generate a warning message and notify the user via their device that "This email may be a scam. Please be careful."

[0925] 4. Example of a prompt statement

[0926] An example of a prompt message for a generative AI model is, "Please analyze the content of this email and assess its likelihood of being a scam."

[0927] In this way, the system can respond quickly and accurately to suspicious communications received by the elderly, thereby preventing them from becoming victims of fraud.

[0928] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0929] Step 1:

[0930] The user inputs the content of suspicious communications received by the elderly person into the terminal. The terminal then sends this input data to the server. The input data is in text format.

[0931] Step 2:

[0932] The server preprocesses the received text data. Specifically, it uses Python's natural language processing libraries, NLTK and spaCy, to normalize the text. This includes removing unnecessary symbols and spaces, and converting words to lowercase. As a result of the preprocessing, text data in a format suitable for analysis is output.

[0933] Step 3:

[0934] The server extracts important keywords and phrases from pre-processed text data. This includes using natural language processing techniques to identify words that may be related to fraud. For example, keywords such as "bank account," "fraudulent use," and "urgent" are extracted. The extracted keywords become input data for evaluating the likelihood of fraud.

[0935] Step 4:

[0936] The server uses a generative AI model to analyze the context based on extracted keywords. The AI ​​model understands the meaning of the entire text and assesses the likelihood of fraud. This assessment determines whether the text is likely to be fraudulent. The evaluation results become input data for generating a warning message.

[0937] Step 5:

[0938] The server generates a warning message based on the evaluation results. If it is determined that there is a high probability of fraud, a warning message is created. For example, a message such as "This email may be a scam. Please be careful." is generated. The generated message is sent to the terminal.

[0939] Step 6:

[0940] The device displays warning messages received from the server to the user. Users can receive warnings visually or audibly. This helps older adults to be alerted to potentially fraudulent communications.

[0941] (Application Example 1)

[0942] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0943] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to issue appropriate warnings. However, conventional methods can lead to delays or incorrect judgments, putting the elderly at risk of becoming victims. To solve this problem, technology is needed that can analyze communication content in real time and determine the potential for fraud with high accuracy.

[0944] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0945] In this invention, the server includes information processing means, evaluation means, and identification means. This makes it possible to analyze received communications in real time and quickly display a warning if there is a possibility of fraud.

[0946] "Information processing means" refers to a device or program that analyzes the content of received communications and performs data processing to determine the possibility of fraud.

[0947] An "evaluation tool" is a device or program that scores the likelihood of fraud based on the analyzed communication content and makes a judgment based on the result.

[0948] "Identification means" refers to a device or program that detects suspicious elements from information provided by elderly individuals and issues a warning based on that detection.

[0949] "Notification means" refers to a device or program that provides a visual or auditory warning to a user in the event of potential fraud.

[0950] "Means of real-time analysis" refers to a device or program that instantly analyzes the content of a communication as soon as it is received and quickly outputs the results.

[0951] The system for implementing this invention involves a server and a terminal working in cooperation. The server is equipped with information processing means, evaluation means, and identification means, and analyzes the communication content received from the terminal. Specifically, the server uses a natural language processing library (e.g., spaCy, NLTK) to tokenize the communication content and extract keywords and phrases. Next, it uses a generative AI model (e.g., OpenAI GPT) to analyze the context of the extracted keywords and phrases and score the likelihood of fraud.

[0952] The device receives analysis results from the server and, if it detects a high probability of fraud, provides the user with a visual or auditory warning. This allows the user to respond quickly to suspicious communications.

[0953] As a concrete example, if a user receives an email with the subject line "Important Notice from the Bank," the device sends its contents to the server. The server extracts keywords such as "bank," "important," and "notice," and analyzes them using a generative AI model. If it is determined that the email is highly likely to be fraudulent, the device displays a warning message saying, "This email may be fraudulent. Do not click on the link."

[0954] An example of a prompt message is: "Please analyze the following email content to determine if it is a scam. Email content: 'Important notice from your bank. Please update your account information.'"

[0955] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0956] Step 1:

[0957] The terminal retrieves the content of communications (phone calls and emails) received by the user. This content is sent to the server as text data. The input is the content of communications received by the user, and the output is the transmission of text data to the server.

[0958] Step 2:

[0959] The server tokenizes the received text data using a natural language processing library (e.g., spaCy, NLTK). This process extracts keywords and phrases from the text data. The input is the text data sent from the terminal, and the output is the extracted keywords and phrases.

[0960] Step 3:

[0961] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the context of the extracted keywords and phrases. This analysis scores the likelihood of fraud. The input is the extracted keywords and phrases, and the output is a score indicating the likelihood of fraud.

[0962] Step 4:

[0963] The server determines whether there is a high probability of fraud based on the scoring results. If it determines that there is a high probability of fraud, it generates a warning message. The input is the score indicating the likelihood of fraud, and the output is the warning message.

[0964] Step 5:

[0965] The terminal notifies the user of warning messages received from the server. These notifications are visual or auditory, warning the user of potential fraud. The input is the warning message from the server, and the output is the warning notification to the user.

[0966] (Example 2)

[0967] Next, we will describe Example 2 of the morphological example. 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."

[0968] When elderly people use the internet, there is a risk of accessing suspicious links or information. Such risks can lead to phishing scams and malware infections, posing a threat to the safety of the elderly. Traditional methods are insufficient to prevent these risks, making it difficult for the elderly to use the internet with peace of mind.

[0969] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0970] In this invention, the server includes information processing means, monitoring means, and notification means. This makes it possible to immediately issue a warning when an elderly person attempts to access suspicious links or information, thereby preventing harm.

[0971] "Information processing means" refers to technology for analyzing suspicious communications received by elderly people and understanding their content.

[0972] "Monitoring measures" refer to technologies that monitor the actions of elderly people in real time and detect suspicious links or information.

[0973] "Notification means" refers to technology that generates warnings for detected suspicious links or information and informs elderly people.

[0974] A "monitoring device" is a device used to monitor the actions of elderly people in real time and has the function of recording the user's behavior.

[0975] "Analysis methods" refer to technologies used to compare suspicious links and information with a database and evaluate their safety.

[0976] "Display means" refers to technologies for displaying warning messages to the elderly visually or audibly.

[0977] This invention is a system for reducing the risk of elderly people accessing suspicious links or information when using the internet. The system includes information processing means, monitoring means, and notification means.

[0978] The server uses information processing tools to analyze suspicious communications received by elderly individuals. This analysis utilizes natural language processing technology to understand the content of the communications. Specifically, the server uses software such as a "natural language processing engine" to analyze the content of emails and messages.

[0979] The terminal uses monitoring devices to monitor the elderly person's actions in real time. Monitoring software installed on the terminal captures links the user attempts to click and URLs of websites they attempt to access, and sends them to a server. Security software is used for this monitoring.

[0980] The server uses analysis tools to compare URLs sent from the terminal against a database of known suspicious links and phishing sites. For example, it uses a "security evaluation API" to assess the security of a URL.

[0981] If suspicious links or information are detected, the server generates a warning message using notification mechanisms. This warning message is sent to the terminal and notified to the elderly visually or audibly through display mechanisms. For example, a message such as "This link is suspicious. Please stop accessing it." might be displayed.

[0982] As a concrete example, consider a scenario where an elderly person opens an email with the subject line "Receive a free gift" and attempts to click on a link within it. The device captures the link and sends it to the server. The server analyzes the link, and if it determines it may be a phishing attempt, it generates a warning message saying, "This link is suspicious. Do not access it," and sends it to the device. The device displays this warning as a pop-up to the user, allowing the user to refrain from clicking the link.

[0983] An example of a prompt to input into the generation AI model would be something like, "Determine whether the link an elderly person is about to click is safe, and generate a warning message if necessary."

[0984] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0985] Step 1:

[0986] Users use the internet.

[0987] Users open emails or browse websites. For example, a user might try to click a link in an email. This action constitutes input.

[0988] Step 2:

[0989] The device monitors the user's actions.

[0990] The terminal uses monitoring software to monitor user actions in real time. Specifically, it captures the URLs of links the user attempts to click and websites they attempt to access. These URLs are then sent to the server as output.

[0991] Step 3:

[0992] The server detects suspicious links and information.

[0993] The server receives the URL sent from the terminal and uses analysis tools to compare it against a database of known suspicious links and phishing sites. This comparison with the database is data processing, and the server evaluates the safety of the URL. The result of this evaluation is the output.

[0994] Step 4:

[0995] The server generates a warning message.

[0996] Based on the analysis results, the server generates a warning message if it detects suspicious links or information. For example, it might create a message such as, "This link is suspicious. Please stop accessing it." This message is then sent to the terminal as output.

[0997] Step 5:

[0998] The device displays a warning to the user.

[0999] The device receives warning messages sent from the server and displays them to the user. Specifically, these are displayed as pop-up notifications or on-screen alerts. This display is the output, alerting the user to the warning. The user can then review the warning and choose a safe course of action.

[1000] (Application Example 2)

[1001] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1002] When elderly people use the internet, there is an increasing risk of fraud and security risks due to accessing suspicious links or information. In this situation, there is a need to provide an environment where elderly people can use the internet with peace of mind.

[1003] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1004] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for issuing warnings and advice regarding the detected suspicious information, a means for scanning links in real time in cooperation with a database for evaluating the safety of links, and a means for displaying warnings based on the evaluation results. This prevents elderly people from accessing suspicious links and information, enabling them to use the internet safely.

[1005] "Elderly people" refers to individuals whose physical or cognitive abilities are likely to decline due to aging.

[1006] "Suspicious communication" refers to means of information transmission, such as phone calls or emails, that contain content that the recipient finds questionable.

[1007] "Generative artificial intelligence methods" refer to technologies that use machine learning and natural language processing to analyze data and make judgments and predictions according to specific purposes.

[1008] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[1009] "Means for detecting suspicious information" refers to technologies for identifying and distinguishing suspicious elements from received information.

[1010] "Means of providing warnings and advice" refers to notification methods that provide users with warnings and guidance on how to act.

[1011] A "database for evaluating link safety" refers to a collection of information used to assess the reliability and safety of links on the internet.

[1012] "Methods for scanning links in real time" refer to technologies that instantly assess the security of a link the moment it is clicked.

[1013] "Means of displaying warnings based on evaluation results" refers to methods for displaying appropriate warnings to users in accordance with the results of link safety evaluations.

[1014] The system for implementing this invention mainly consists of a server and a user terminal. The server uses generative artificial intelligence to analyze suspicious communications received by elderly people and determine if they are fraudulent. Specifically, the server utilizes natural language processing technology to analyze the content of received emails and phone calls. This allows it to identify potentially fraudulent information and issue warnings to the user.

[1015] The user's device has an application installed to evaluate the security of links. This application scans links in real time as they are clicked and evaluates their security in conjunction with a database. Based on the evaluation results, the device displays a visual or audible warning to the user. This helps prevent elderly people from accessing suspicious links.

[1016] For example, if a user attempts to click a link in an email, and the system determines that the link may lead to a phishing site, it will display a warning message such as, "This link is not secure. Please stop accessing it." An example of a prompt message generated using an AI model would be, "Is this link secure? Please evaluate the link's URL."

[1017] In this way, the server and the user's terminal work together to provide an environment that allows elderly people to use the internet safely.

[1018] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1019] Step 1:

[1020] The user receives an email. The email content is displayed on the device. The user attempts to click a link in the email.

[1021] Step 2:

[1022] The device retrieves the URL of the clicked link. It then sends the retrieved URL to the server.

[1023] Step 3:

[1024] The server analyzes the received URL using a generation AI model. Specifically, it uses natural language processing technology to evaluate the structure and content of the URL and determine its potential for fraud. As a result of the analysis, it outputs an evaluation of whether the URL is safe or not.

[1025] Step 4:

[1026] The server sends the URL evaluation result to the device. The evaluation result includes information on whether the link is safe or suspicious.

[1027] Step 5:

[1028] The device displays a warning to the user based on the evaluation results received from the server. Specifically, if a link is deemed suspicious, it displays a warning message such as, "This link is not secure. Please stop accessing it."

[1029] Step 6:

[1030] The user reviews the warning from their device and decides whether to cancel access to the link. If the user cancels access, the device blocks the connection to the link.

[1031] This series of processes prevents elderly people from accessing suspicious links, enabling them to use the internet safely.

[1032] (Example 3)

[1033] Next, we will describe Embodiment 3 of Embodiment Example 3. 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."

[1034] There is a need to provide effective warnings and advice to prevent elderly people from becoming victims of fraud by addressing suspicious communications and links they receive. However, conventional methods are insufficient in detecting suspicious links and providing warnings, and the risk of elderly people becoming victims of fraud still remains. To solve this problem, notifications in a format that elderly people can easily understand and more accurate detection of suspicious links are necessary.

[1035] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[1036] In this invention, the server includes generative artificial intelligence means for analyzing suspicious communications received by elderly people, means for determining the possibility of fraud, and means for analyzing suspicious links and evaluating their safety using a machine learning model. This makes it possible to provide elderly people with visually or audibly effective warnings and advice, thereby preventing them from becoming victims of fraud.

[1037] "Generative artificial intelligence means" refers to artificial intelligence technology used to analyze suspicious communications received by elderly people and determine the possibility of fraud.

[1038] "Means of determining the possibility of fraud" refers to the process of analyzing the content of received communications or links and evaluating whether they are related to fraudulent activity.

[1039] "Means for detecting suspicious links and information" refers to technologies that identify suspicious elements within communications and links and issue warnings to users.

[1040] "Means of providing warnings and advice" refers to methods of providing users with messages or guidelines to alert them to suspicious links or information that have been detected.

[1041] "Methods for evaluating safety using machine learning models" refers to the process of analyzing links and information using algorithms that have learned from past data in order to evaluate their safety.

[1042] "Means of providing visual or auditory notification" refers to methods of communicating warnings about suspicious links or information to users through visual displays or audio.

[1043] As an embodiment of this invention, the following system is constructed.

[1044] The server uses a generative AI model to analyze suspicious communications received by elderly individuals. This involves using natural language processing libraries and machine learning frameworks. Specifically, it utilizes software such as "spaCy" and "TensorFlow" to analyze communication content and detect suspicious links and information. The server then uses a machine learning model to evaluate the safety of the detected suspicious links and determine their potential for fraud.

[1045] The device receives warning messages and advice sent from the server and notifies the elderly visually or audibly. Specifically, it uses pop-up messages and audio alerts to warn users not to click on suspicious links.

[1046] The user can initiate the system's operation by entering prompts for the generated AI model. For example, the following prompts might be used:

[1047] "Please create a program that can determine if links in emails received by elderly people are safe and display a warning if necessary."

[1048] Based on this prompt, the generating AI model creates a program to detect suspicious links and issue warnings. This enables elderly people to take concrete actions to prevent themselves from becoming victims of fraud. The flow of the specific processing in Example 3 will be explained using Figure 15.

[1049] Step 1:

[1050] The user inputs a prompt into the generative AI model. For example, the prompt might be, "Create a program that determines whether a link in an email received by an elderly person is safe and displays a warning if necessary." Based on this input, the generative AI model generates a program to detect suspicious links.

[1051] Step 2:

[1052] The server uses the generated program to analyze emails and communications received by elderly individuals. The input includes email content and link information. The server uses the natural language processing library "spaCy" to analyze the email content and extract links and domain information. The output provides potential suspicious links.

[1053] Step 3:

[1054] The server evaluates the extracted link information using a machine learning model. The input includes the suspicious link candidates obtained in step 2. The server uses TensorFlow to assess the safety of the links and determine their likelihood of being fraudulent. The output is a list of suspicious links.

[1055] Step 4:

[1056] The server generates warning messages and advice for links it deems suspicious. The input includes the list of suspicious links obtained in step 3. The server creates a warning message for the user, such as "This link is suspicious. Do not click it." The output is the generated warning message.

[1057] Step 5:

[1058] The server sends the generated warning message to the terminal. The input includes the warning message generated in step 4. The server prepares to send the warning message to the terminal and notify the user. The output is that the message has been sent to the terminal.

[1059] Step 6:

[1060] The terminal displays warning messages received from the server to the user. Input includes the warning messages sent from the server. The terminal provides visual or auditory notifications to the elderly using pop-up messages or audio alerts. Output is the completion of the warning to the user.

[1061] (Application Example 3)

[1062] Next, we will describe application example 3 of form example 3. 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."

[1063] One challenge when elderly people use digital communications is preventing them from becoming victims of fraud or malicious links. In particular, elderly people have difficulty distinguishing suspicious links and information, which increases their risk of becoming victims of fraud. There is a need to improve this situation and provide an environment where elderly people can use digital communications with peace of mind.

[1064] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[1065] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for scanning communication links in real time and issuing a warning when a suspicious link is detected, and a means for issuing warnings through visual notifications or audio alerts. This makes it possible for elderly people to easily identify suspicious links and information and prevent themselves from becoming victims of fraud.

[1066] "Elderly people" generally refers to older individuals who require support in using digital technologies.

[1067] "Suspicious communications" refers to means of information transmission such as phone calls, emails, and messages that may be involved in fraud or illegal activities.

[1068] "Generative artificial intelligence" refers to artificial intelligence technologies used to analyze data and detect specific patterns or anomalies.

[1069] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[1070] "Means for detecting suspicious information" refers to technologies for identifying fraudulent or questionable elements from received information.

[1071] "Real-time scanning" refers to a technology that analyzes information instantly upon receipt and provides the results.

[1072] "Warning mechanisms" refer to technologies that provide notifications and alerts to inform users of potential dangers.

[1073] "Visual notifications" refer to means of conveying information through visual means, such as messages and pop-ups displayed on the screen.

[1074] "Voice alert" refers to a warning method that uses sound to draw the user's attention.

[1075] The system for implementing this invention mainly consists of a server and a terminal. The server uses generative artificial intelligence to analyze suspicious communications received by elderly people and determine if they are fraudulent. Specifically, the server utilizes natural language processing technology to analyze the content of communications. This allows it to detect suspicious information and perform real-time scans.

[1076] The device scans the links of communications received by the user in real time and issues a warning if it detects a suspicious link. The warning is provided to the user through visual notifications and audio alerts. Visual notifications appear as pop-ups on the device screen, and audio alerts are generated using a Text-to-Speech (TTS) engine.

[1077] As a concrete example, a device automatically scans a link received by a user via SMS and displays a warning message saying, "This link may lead to a phishing site. Do not click it." In this process, the server uses the Google Safe Browsing API to evaluate the safety of the link.

[1078] An example of a prompt to input into a generating AI model is, "Is this link safe? If it may be a phishing site, please display a warning." Using this prompt, the system can provide appropriate warnings to the user.

[1079] The flow of the specific processing in Application Example 3 will be explained using Figure 16.

[1080] Step 1:

[1081] The user receives the communication on their device. The device extracts links and text data to analyze the content of the received communication. The input is the received communication data, and the output is the links and text data to be analyzed.

[1082] Step 2:

[1083] The device sends extracted links to the server for real-time scanning. The server uses the Google Safe Browsing API to evaluate the safety of the links. The input is the links to be analyzed, and the output is the evaluation result regarding the safety of the links.

[1084] Step 3:

[1085] The server uses generative artificial intelligence to analyze the text data of received communications using natural language processing. This allows it to determine whether the communications are potentially fraudulent. The input is the text data of the communications, and the output is the evaluation result regarding the likelihood of fraud.

[1086] Step 4:

[1087] The server integrates the link safety assessment results and the text data analysis results, and generates a warning if suspicious information is detected. The input is the link safety assessment results and the fraud potential assessment results, and the output is a warning message.

[1088] Step 5:

[1089] The terminal notifies the user of warning messages received from the server. It displays a pop-up on the screen as a visual notification and plays the warning as an audio alert using a Text-to-Speech (TTS) engine. The input is the warning message, and the output is the warning notification to the user.

[1090] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1091] "Example of form 1"

[1092] One embodiment of the present invention involves a system that combines a generative AI for analyzing suspicious phone calls and emails received by elderly people, means for determining the possibility of fraud, means for detecting suspicious links and information provided by elderly people, and means for providing warnings and advice regarding detected suspicious links and information, with an emotion engine. This emotion engine analyzes the emotions of elderly people from their tone of voice and facial expressions, and incorporates the results as part of the fraud judgment. Specifically, it analyzes the tone and speed of the elderly person's voice when they speak on the phone, or changes in facial expressions read from video calls or photographs, and reflects the results in the fraud judgment.

[1093] "Example of form 2"

[1094] Furthermore, the emotional engine adjusts how warnings and advice are delivered based on the emotions of the elderly. For example, if it determines that an elderly person is feeling anger or confusion, it uses that information to adjust the tone and expression of warnings and advice, providing information in a way that is easy for the elderly person to understand. This makes it possible to provide appropriate advice and protect elderly people from fraud even when they are emotionally charged.

[1095] "Example of form 3"

[1096] Furthermore, the emotion engine can track changes in the emotions of older adults over the long term and incorporate that information into fraud detection. For example, if an older adult is consistently judged to be experiencing anxiety or fear, this information can be used to increase the likelihood of fraud and encourage more careful monitoring. This allows for the use of changes in the emotions of older adults in fraud detection, resulting in more accurate fraud detection.

[1097] The following describes the processing flow for each example of the form.

[1098] "Example of form 1"

[1099] Step 1: A generative AI analyzes suspicious phone calls and emails received by elderly individuals.

[1100] Step 2: Based on the analysis results, determine the likelihood of fraud.

[1101] Step 3: Detect suspicious links and information provided by elderly individuals.

[1102] Step 4: The emotion engine analyzes the emotions of the elderly person from their tone of voice and facial expressions.

[1103] Step 5: Use the analysis results as part of the fraud determination.

[1104] "Example of form 2"

[1105] Step 1: The emotion engine analyzes the emotions of the elderly.

[1106] Step 2: Based on the analysis results, adjust how warnings and advice are provided.

[1107] Step 3: Provide adjusted warnings and advice to the elderly.

[1108] "Example of form 3"

[1109] Step 1: The emotion engine tracks changes in older adults' emotions over the long term.

[1110] Step 2: Use the tracking results to determine if it's a scam.

[1111] Step 3: Based on the results, increase the likelihood of fraud and monitor more carefully.

[1112] (Example 1)

[1113] Next, we will describe Embodiment 1 of Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1114] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to provide appropriate warnings and advice. However, conventional technologies are insufficient in identifying fraud or may delay warnings, which presents a challenge as the elderly are at a high risk of becoming victims of fraud.

[1115] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1116] In this invention, the server includes information processing means, evaluation means, identification means, sentiment analysis means, and notification means. This makes it possible to quickly analyze suspicious communications received by the elderly, determine the possibility of fraud with high accuracy, and provide appropriate warnings and advice.

[1117] "Information processing means" refers to technical means for analyzing suspicious communications received by elderly people, and has the function of analyzing the content of communications using natural language processing technology.

[1118] "Evaluation means" refers to technical means for determining the possibility of fraud based on the analyzed communication content, and has the function of scoring the possibility of fraud using a generative AI model.

[1119] "Identification means" refers to technical means for detecting suspicious information provided by elderly individuals, and includes the function of evaluating the security of links and information.

[1120] "Emotional analysis means" refers to technical means for analyzing the emotions of elderly people and reflecting the results in fraud detection. These means have the function of analyzing emotions using voice recognition technology and image analysis technology.

[1121] "Notification means" refers to technical means for issuing warnings or advice regarding detected suspicious information, and has the function of providing visual or auditory notifications to the elderly.

[1122] This invention is a system for analyzing suspicious communications received by elderly people and determining the possibility of fraud. The system includes information processing means, evaluation means, identification means, sentiment analysis means, and notification means.

[1123] The server uses software such as Python and TensorFlow as information processing tools to analyze the content of suspicious communications received by elderly individuals using natural language processing techniques. This makes it possible to extract keywords and phrases contained in the communications and analyze the context.

[1124] As an evaluation method, the server uses a generative AI model to score the likelihood of fraud based on the analyzed data. If it is determined that there is a high probability of fraud, the process proceeds to the next step.

[1125] The identification mechanism works to detect suspicious links and information provided by users. The server evaluates the safety of the link and issues a warning if it determines that it is dangerous.

[1126] As a means of emotion analysis, the server uses speech recognition and image analysis technologies to analyze the tone and speed of the elderly person's voice, as well as changes in their facial expressions. This allows the server to determine whether the elderly person is feeling anxious or tense, and this information is then used to determine if they are a fraud suspect.

[1127] As a notification method, the server sends warnings and advice to the terminal. The terminal provides elderly users with visual or auditory notifications and displays specific advice. For example, it may display a message such as, "This call may be a scam. Do not provide any personal information."

[1128] For example, if a user receives a suspicious email, their device forwards the email to the server. The server analyzes the email's content, and if it detects phrases such as "verify bank account information," it warns the user, "This email may be a scam. Do not click the link."

[1129] Examples of prompt messages include the following:

[1130] "Could this phone call be a scam?"

[1131] "Are the links in the email safe?"

[1132] "Does this tone of voice indicate anxiety?"

[1133] In this way, the system analyzes suspicious information received by the elderly, determines the possibility of fraud, and issues a warning if necessary.

[1134] The flow of the specific processing in Example 1 will be explained using Figure 17.

[1135] Step 1:

[1136] When a user receives a suspicious phone call or email, they input the content into their device. The device then sends the audio or text data to the server. Specifically, data is collected using the smartphone's recording function or email app. The input is audio files or text data, and the output is data transfer to the server.

[1137] Step 2:

[1138] The server analyzes the received data using natural language processing techniques. Using Python and TensorFlow, it extracts keywords and phrases from phone calls and emails and analyzes the context. Input is audio files or text data, and output is a list of the analyzed keywords and phrases.

[1139] Step 3:

[1140] The server determines the likelihood of fraud based on the analysis results. Using a generative AI model, it scores the likelihood of fraud from detected keyword and phrase combinations. The input is a list of analyzed keywords and phrases, and the output is a score indicating the likelihood of fraud.

[1141] Step 4:

[1142] The server analyzes the user's emotions using an emotion engine. It uses speech recognition technology to analyze the tone and speed of the user's voice, and image analysis technology to read changes in facial expressions from video calls and photos. The input is audio data and image data, and the output is the result of the emotion analysis.

[1143] Step 5:

[1144] The server will issue a warning to the user if it determines that the call is likely to be a scam. It will send a notification to the device, displaying specific advice and warnings. The input is a score indicating the likelihood of fraud and the result of sentiment analysis, and the output is a warning message to the user. Specifically, it will display a message such as, "This call may be a scam. Do not provide any personal information."

[1145] (Application Example 1)

[1146] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1147] There is a need to quickly and accurately determine the possibility of fraud in suspicious communications received by the elderly and to provide appropriate warnings and advice. However, conventional technologies have limitations in analyzing the content of communications and do not take into account the emotional state of the elderly when making fraud judgments, making it difficult to prevent fraud.

[1148] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1149] In this invention, the server includes a generative information processing means for analyzing suspicious communications received by an elderly person, an information processing means for determining the possibility of fraud, and an information processing means for analyzing the elderly person's voice and facial expressions to detect emotions. This makes it possible to make fraud judgments that take into account the emotional state of the elderly person.

[1150] "Elderly people" refers to individuals whose physical or cognitive abilities are likely to decline as a result of aging.

[1151] "Suspicious communication" refers to means of transmitting information, such as phone calls or emails, that may have questionable intentions or content directed at the recipient.

[1152] "Generative information processing means" refers to information processing technology that analyzes received information and makes decisions or generates information according to a specific purpose.

[1153] "Information processing means for determining the possibility of fraud" refers to information processing technology that analyzes received information and evaluates whether that information is related to fraudulent activity.

[1154] "Information processing means for detecting suspicious information" refers to information processing technology for identifying and detecting suspicious elements from received information.

[1155] "Information processing means for issuing warnings and advice" refers to information processing technology that provides recipients with warnings and guidance on action based on detected suspicious information.

[1156] "Information processing means for analyzing voice and facial expressions to detect emotions" refers to information processing technology that analyzes the voice and facial expressions of a recipient to identify their emotional state.

[1157] "Information processing means that reflect the results of emotional analysis in fraud judgment" refers to information processing technology that incorporates analyzed emotional information into the process of evaluating the likelihood of fraud.

[1158] The system for implementing this invention mainly consists of a server and a terminal. The server is equipped with generative information processing means for analyzing suspicious communications received by elderly people, and analyzes the content of the communications using natural language processing technology. Specifically, it uses natural language processing libraries (e.g., spaCy or NLTK) to extract keywords and phrases from the communications and evaluate the possibility of fraud.

[1159] Furthermore, the server is equipped with information processing capabilities to analyze the voice and facial expressions of elderly individuals and detect their emotions. This involves using emotion analysis libraries (such as OpenCV or DeepFace) to analyze audio and video data acquired through the device's camera and microphone. The analysis results are used to determine if a fraud is likely, and if there is a high probability of fraud, warnings and advice are displayed on the device.

[1160] The device sends communications received by the user to a server and receives analysis results from the server. The user receives visual or auditory notifications on the device to be alerted to potentially fraudulent communications.

[1161] For example, if a user receives an email titled "Important Notice from the Bank," the device sends its contents to a server. The server uses natural language processing to analyze keywords such as "bank," "important," and "notice" to determine if it is a scam. Additionally, when a user receives a phone call, the device's camera analyzes their facial expressions to detect changes in emotion. If it is determined that the email is likely to be a scam, a warning message such as "This email may be a scam. Do not click the link." is displayed on the device.

[1162] An example of a prompt to input into the generating AI model is: "Analyze the contents of this email and determine if it is a scam. The email contains the following: 'Important notice from the bank.'"

[1163] The flow of a specific process in Application Example 1 will be explained using Figure 18.

[1164] Step 1:

[1165] The terminal sends suspicious communications (phone calls and emails) received by the user to the server. The input is the content of the communication, and the output is the data to be sent to the server. The terminal converts the communication content into text format and generates data packets for transmission to the server.

[1166] Step 2:

[1167] The server analyzes the received communication content using generative information processing means. The input is the communication content sent from the terminal, and the output is the analysis result. The server uses natural language processing libraries (e.g., spaCy or NLTK) to extract keywords and phrases from the communication content and evaluate the likelihood of fraud.

[1168] Step 3:

[1169] The server analyzes the user's voice and facial expressions to detect emotions. The input is audio and video data sent from the terminal, and the output is the emotion analysis result. The server uses emotion analysis libraries (e.g., OpenCV or DeepFace) to analyze changes in voice tone and facial expressions.

[1170] Step 4:

[1171] The server integrates the results of communication content analysis and sentiment analysis to determine the likelihood of fraud. The input is the results of communication content analysis and sentiment analysis, and the output is the evaluation of the likelihood of fraud. The server uses a generative AI model to make a comprehensive judgment on the likelihood of fraud.

[1172] Step 5:

[1173] The server sends a warning or advice to the terminal if it determines that there is a high probability of fraud. The input is the result of the fraud probability assessment, and the output is the warning message. The server generates the warning message and creates a data packet to send to the terminal.

[1174] Step 6:

[1175] The terminal notifies the user of warnings and advice received from the server. Input is the warning message from the server, and output is the notification to the user. The terminal displays or plays the warning message to the user visually or audibly.

[1176] (Example 2)

[1177] Next, we will describe Example 2 of the morphological example. 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."

[1178] When elderly people use the internet, there is a problem in that they are at a higher risk of becoming victims of fraud because they cannot respond appropriately to suspicious communications and information. In addition, they may not be able to correctly understand information because they are not provided with appropriate warnings and advice that are tailored to their emotional state.

[1179] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1180] In this invention, the server includes a generative AI means for analyzing suspicious communications received by elderly people, means for determining the possibility of fraud, and means for detecting suspicious information provided by elderly people. This enables elderly people to respond quickly and appropriately to suspicious information and prevent them from becoming victims of fraud.

[1181] "Generative AI methods" refer to artificial intelligence technologies used to analyze suspicious communications received by elderly people and to understand their content.

[1182] "Means for determining the possibility of fraud" refers to algorithms or processes that evaluate whether a communication is fraudulent based on its analyzed content.

[1183] "Methods for detecting suspicious information" refer to technologies for identifying information that exhibits unusual patterns or has known risks from information provided by elderly individuals.

[1184] "Warning measures" refer to methods of providing visual or auditory notifications to alert older adults to detected suspicious information.

[1185] "Methods for analyzing emotional states" refer to techniques for analyzing the tone of voice and facial expressions of elderly people to determine their emotional state.

[1186] "Means of adjusting the content of warnings" refers to methods of modifying the tone and expression of warnings and advice to make them easier for older adults to understand, based on their analyzed emotional state.

[1187] This invention is a system that enables elderly people to appropriately respond to suspicious communications and information when using the internet. The system consists of three components: a server, a terminal, and a user.

[1188] The server uses generative AI methods to analyze suspicious communications received by elderly individuals. This analysis employs natural language processing techniques to understand the content of the communications and determine if they are fraudulent. The server runs algorithms that cross-reference information with a database to identify information that poses a known risk.

[1189] The device uses voice recognition software and a facial recognition camera to analyze the emotional state of elderly individuals. This allows the device to analyze the tone of voice and facial expressions of elderly individuals in real time to determine their emotional state. Based on this emotional state, the device adjusts the content of warnings and advice, providing information in a way that is easy for the elderly to understand.

[1190] Elderly users receive warnings and advice from the device and take appropriate action. For example, when an elderly person inputs the prompt "Is it safe to click this link?" into the AI ​​model, the system evaluates the safety of the link and returns appropriate advice. This allows elderly people to use the internet with peace of mind.

[1191] The flow of the specific processing in Example 2 will be explained using Figure 19.

[1192] Step 1:

[1193] The server monitors communications received by elderly individuals and retrieves emails and website links as input. Using generative AI, the server analyzes the content of these communications using natural language processing techniques. This analysis generates data to determine the likelihood of a communication being fraudulent. The output provides the analysis results and a flag indicating whether the communication is potentially fraudulent.

[1194] Step 2:

[1195] The server detects suspicious information based on the analysis results. It uses a database of analyzed communication content and information with known risks as input. The server compares these and runs algorithms to identify suspicious links and information. A list of suspicious information is generated as output.

[1196] Step 3:

[1197] The terminal receives a list of suspicious information sent from the server and uses it as input. The terminal uses an emotion engine to analyze the emotional state of elderly people. Specifically, it uses speech recognition software and a facial recognition camera to analyze the tone of voice and facial expressions of elderly people in real time. The output is an evaluation result of their emotional state.

[1198] Step 4:

[1199] The device adjusts the content of warnings and advice based on the results of an emotional state assessment. It uses a list of suspicious information and the results of the emotional state assessment as input. The device modifies the tone and expression of warnings and advice to make them easily understandable to older adults. The adjusted warnings and advice are generated as output and provided to older adults as visual or auditory notifications.

[1200] Step 5:

[1201] The elderly user receives warnings and advice from the device and takes appropriate action. Specifically, the elderly user can input a prompt message such as "Is it safe to click this link?" into the AI ​​model. Based on this prompt, the system evaluates the safety of the link and provides feedback to the user. As output, the elderly user is presented with advice regarding the safety of the link.

[1202] (Application Example 2)

[1203] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1204] Fraud cases involving elderly people accessing suspicious links and information online are on the rise. In this situation, there is a need to provide an environment where elderly people can use the internet with peace of mind. Furthermore, it is important to provide appropriate warnings and advice that take into account the feelings of the elderly.

[1205] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1206] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, a means for detecting suspicious information provided by elderly people, a means for issuing warnings and advice regarding the detected suspicious information, an emotion analysis means for analyzing the emotions of elderly people and adjusting the expression of warnings and advice, and a means for verifying the safety of suspicious information using an external database. This makes it possible to reduce the risk of elderly people accessing suspicious information and to provide appropriate warnings and advice that take their emotions into consideration.

[1207] "Elderly people" refers to individuals who are considered to be prone to a decline in physical or cognitive abilities as a result of aging.

[1208] "Suspicious communication" refers to means of information transmission such as phone calls and emails that may have been sent to the recipient with an improper purpose.

[1209] "Generative artificial intelligence means" refers to methods that use artificial intelligence technology to analyze data and make judgments or predictions according to specific purposes.

[1210] "Means of determining the possibility of fraud" refers to algorithms or processes used to evaluate whether received information is related to fraudulent activity.

[1211] "Means of detecting suspicious information" refers to methods for identifying information that is deemed unreliable or dangerous from among the information provided on the internet.

[1212] "Means of providing warnings and advice" refers to means of generating notifications and messages to draw attention to users or provide guidance on how to act.

[1213] "Emotional analysis tools" refer to methods for analyzing a user's emotional state and taking appropriate action based on the results.

[1214] "Methods of verification using external databases" refer to methods of verifying information by referring to external databases in order to evaluate the safety and reliability of information on the internet.

[1215] The system for implementing this invention mainly consists of a server and a user terminal. The server uses generative artificial intelligence means to analyze suspicious communications received by elderly people. Specifically, a program developed using Python utilizes the Google Cloud Natural Language API to analyze the content of the communications using natural language processing. This determines whether the communications are likely to be fraudulent.

[1216] On the user's device, it is implemented as an internet browser extension to detect suspicious information. It uses the VirusTotal API to verify the safety of links and information. For detected suspicious information, a notification is displayed to provide warnings and advice. This notification is delivered in an appropriate tone, based on an analysis of the user's emotional state using sentiment analysis tools. The Google Cloud Natural Language API is used for sentiment analysis.

[1217] For example, if an elderly person attempts to click on a link in an email, the system may detect that the link is suspicious. If the sentiment analysis system determines that the user is confused, a gentle warning message will be displayed, such as, "This link may not be safe. Please check the details."

[1218] An example of a prompt message is, "The user is about to click a link. Please verify the safety of this link and display a warning appropriate to the user's sentiment." This prompt message allows the system to take appropriate action and protect elderly users from suspicious information.

[1219] The flow of a specific process in Application Example 2 will be explained using Figure 20.

[1220] Step 1:

[1221] The user attempts to click a link on their device. The device retrieves the URL of the clicked link. This URL becomes the input data.

[1222] Step 2:

[1223] The device sends the retrieved URL to the VirusTotal API for security verification. The VirusTotal API evaluates the URL's security and outputs the result. The output indicates whether the URL is safe or suspicious.

[1224] Step 3:

[1225] The server receives evaluation results from the VirusTotal API, and if it determines that the URL is suspicious, it collects sentiment data from the user's device to analyze the user's sentiment. This data includes the user's recent activity history and input content.

[1226] Step 4:

[1227] The server sends the collected emotion data to the Google Cloud Natural Language API for emotion analysis. The input data is information about the user's emotions, and the output is the analysis result showing the user's emotional state.

[1228] Step 5:

[1229] The server generates a warning message in an appropriate tone based on the sentiment analysis results. It uses a generative AI model to create prompts that respond to the user's emotions. For example, if the user is confused, it might generate a message such as, "This link may not be safe. Please check the details."

[1230] Step 6:

[1231] The device displays warning messages received from the server to the user. This allows the user to receive warnings about suspicious links and take appropriate action.

[1232] (Example 3)

[1233] Next, we will describe Embodiment 3 of Embodiment Example 3. 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."

[1234] Reducing the risk of elderly people becoming victims of fraud is crucial. In particular, it is important to appropriately identify suspicious communications and information and issue prompt warnings. Improving the accuracy of fraud detection, taking into account the emotional changes of the elderly, is also a challenge.

[1235] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 3 is realized by the following means.

[1236] In this invention, the server includes information processing means, judgment means, detection means, notification means, and emotion analysis means. This makes it possible to analyze suspicious communications received by elderly people, assess the possibility of fraud, and provide appropriate warnings and advice. Furthermore, by tracking changes in the elderly person's emotions and reflecting them in fraud judgments, the accuracy of fraud detection can be improved.

[1237] "Information processing means" refers to technology for analyzing suspicious communications received by elderly people and understanding their content.

[1238] "Decision-making tools" are techniques used to evaluate the likelihood of fraud based on analyzed information.

[1239] "Detection means" refers to technology for identifying and distinguishing suspicious information provided by elderly individuals.

[1240] "Notification methods" refer to technologies used to communicate warnings and advice to the elderly regarding detected suspicious information.

[1241] "Emotional analysis methods" refer to technologies that track changes in the emotions of elderly people and use that information to make fraud judgments.

[1242] This invention is a fraud detection system for the elderly, realized through the respective roles of the server, terminal, and user. The server uses information processing means, judgment means, detection means, notification means, and sentiment analysis means to analyze suspicious communications received by the elderly and evaluate the possibility of fraud.

[1243] Specifically, the server uses a generative AI model to scan links and information contained in emails and messages to detect suspicious elements. For example, it analyzes the URL of a link and compares it against a list of known fraudulent sites to determine the likelihood of fraud. It also uses an emotion engine to collect user emotion data and track changes in their emotions. This allows the system to incorporate information about when a user is feeling anxious or fearful into its fraud detection.

[1244] The device provides visual or auditory notifications to the elderly based on instructions from the server. For example, if a user attempts to click a suspicious link, the device displays a warning such as, "This link may be a scam. Do not click it." It also provides advice based on emotional data, such as, "You seem to have been feeling anxious recently. Be careful of scams."

[1245] Users receive notifications from their devices and take specific actions. For example, they may be prompted not to click on a link they are warned about or to delete a suspicious email.

[1246] An example of a prompt message is, "Please determine if the link in this email is safe." The generating AI model then analyzes the link's safety and notifies the user of the result. Figure 21 illustrates the specific processing flow in Example 3.

[1247] Step 1:

[1248] Users generate data through everyday actions. Inputs include receiving emails, browsing websites, and sending and receiving messages. This data is sent to a server. The server collects this data and prepares it for analysis. Specifically, it stores email content and link information in a database.

[1249] Step 2:

[1250] The server analyzes the collected data. The input is data from emails and messages sent by users. The server uses a generative AI model to scan links and information contained in emails and messages, detecting suspicious elements. Specifically, it analyzes the URLs of links and compares them against a list of known fraudulent websites. The output is a list of suspicious links and information.

[1251] Step 3:

[1252] The server uses an emotion engine to collect user emotion data. Input includes user behavior and input data while using the device. The server analyzes this data to infer changes in the user's emotions. Specifically, if the user frequently exhibits anxiety-inducing behaviors, this data is recorded. The output is a report on changes in emotions.

[1253] Step 4:

[1254] The server integrates analysis results and sentiment data to determine the likelihood of fraud. Inputs include a list of suspicious links and information, and reports on changes in sentiment. Based on this data, the server assesses the risk of fraud and generates warnings as needed. The output is a warning message.

[1255] Step 5:

[1256] The device provides visual or auditory notifications to the user based on instructions from the server. The input is a warning message from the server. If the user attempts to click a suspicious link, the device will display a warning such as, "This link may be fraudulent. Do not click it." Specifically, this involves displaying a large red warning message on the screen. The output is a warning notification to the user.

[1257] Step 6:

[1258] The user receives a notification from their device and takes specific action. The input is a warning notification from the device. The user is prompted to take action such as not clicking the warned link or deleting a suspicious email. The specific action is to choose an action to avoid the risk of fraud. The output is that the user's safe actions are promoted.

[1259] (Application Example 3)

[1260] Next, we will describe application example 3 of form example 3. 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."

[1261] There is a need to reduce the risk of elderly people becoming victims of fraud. However, conventional methods have difficulty in appropriately detecting and warning about suspicious communications and information, and fraud judgments do not take into account the emotional changes of the elderly, resulting in insufficient accuracy in fraud prevention.

[1262] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[1263] In this invention, the server includes a generative artificial intelligence means for analyzing suspicious communications received by elderly people, a means for determining the possibility of fraud, and an emotion analysis means for tracking changes in the elderly person's emotions and reflecting them in the fraud determination. This reduces the risk of fraud against elderly people and enables more accurate fraud prevention.

[1264] "Generative artificial intelligence methods" refer to methods that use artificial intelligence technology to analyze suspicious communications received by elderly people and determine the possibility of fraud.

[1265] "Means of determining the possibility of fraud" refers to methods of evaluating whether the analyzed information is related to fraud.

[1266] "Means for detecting suspicious information" refers to methods for identifying suspicious elements in information provided by elderly individuals and issuing warnings.

[1267] "Means of providing warnings or advice" refers to means of visually or audibly drawing the attention of elderly individuals to any suspicious information that has been detected.

[1268] "Emotional analysis methods" refer to techniques for tracking changes in the emotions of elderly individuals and using that information to inform fraud detection.

[1269] "Notification means" refers to methods for sending visual or auditory warnings to elderly people.

[1270] The system for implementing this invention mainly consists of a server and a terminal. The server is equipped with generative artificial intelligence means for analyzing suspicious communications received by elderly people. This means analyzes the content of communications using natural language processing technology and determines the possibility of fraud. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to build a model for analyzing the content of communications.

[1271] The devices in question are smartphones, tablets, and other devices used by the elderly, and these devices have notification mechanisms implemented. These notification mechanisms issue visual or auditory warnings based on analysis results. For example, visual warnings can be displayed using Android Notifications or iOS Push Notifications, and voice alerts can be provided using Google Assistant or Siri.

[1272] Furthermore, the server is equipped with emotion analysis capabilities to track changes in the emotions of elderly individuals. This capability collects voice and facial expression data from elderly individuals and uses an AI model to analyze changes in their emotions. This makes it possible to incorporate emotional information into fraud detection.

[1273] For example, if an elderly person opens an email on their smartphone and it contains a suspicious link, the device will issue a voice warning saying, "This link may be a scam. Do not click it," and display a red warning message on the screen.

[1274] An example of a prompt message would be, "Is this link safe? If it may be a scam, please generate a warning message." This would enable older adults to take concrete actions to protect themselves from fraud.

[1275] The flow of the specific processing in Application Example 3 will be explained using Figure 22.

[1276] Step 1:

[1277] The server receives communication data received by elderly individuals. The input consists of email and phone call content, which is then passed to a generative artificial intelligence system. The server analyzes the communication content using natural language processing techniques to determine the likelihood of fraud. The output is the result of the determination of whether or not fraud is likely.

[1278] Step 2:

[1279] The server detects suspicious information based on the analysis results. The input is the judgment result obtained in step 1, which is used to identify suspicious links and information. The server lists the identified suspicious information and sends it to the terminal. The output is a list of suspicious information.

[1280] Step 3:

[1281] The terminal issues a warning to the user based on a list of suspicious information received from the server. The input is a list of suspicious information, which is used to provide visual or auditory warnings. The terminal displays a warning message on the screen and uses a voice assistant to draw attention. The output is the notification of the warning to the user.

[1282] Step 4:

[1283] The server collects emotional data from elderly individuals and analyzes it using emotion analysis tools. Inputs include voice and facial expression data, which are then analyzed by an AI model. The server tracks emotional changes and incorporates them into fraud detection. The output is the result of the emotion analysis.

[1284] Step 5:

[1285] The server improves the accuracy of its fraud detection based on the sentiment analysis results. The input consists of the sentiment analysis results and the judgment results from Step 1, which are integrated to assess the final fraud risk. The server sends the assessment results to the terminal and strengthens warnings as needed. The output is the final fraud risk assessment.

[1286] (Other examples)

[1287] Since this is the same as the specific processing described in the other embodiments of the first embodiment above, the explanation will be omitted.

[1288] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1289] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1290] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[1291] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1292] [Fourth Embodiment]

[1293] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1294] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1295] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1296] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1297] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1298] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1299] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1300] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1301] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1302] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1303] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1304] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1305] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[1306] "Example of form 1"

[1307] One embodiment of the present invention involves a generative AI system that analyzes suspicious phone calls and emails received by elderly individuals. This generative AI uses natural language processing technology to analyze the content of phone calls and emails and determine the likelihood of them being scams. Specifically, it analyzes keywords, phrases, and context contained in the content of phone calls and emails, and if it determines that there is a high probability of fraud, it warns the elderly individual with that information.

[1308] "Example of form 2"

[1309] Furthermore, there are methods to detect suspicious links and information provided by elderly individuals. These methods monitor the actions of elderly individuals when they use the internet, and if they attempt to provide suspicious links or information, they detect the action and warn the elderly individual. Specifically, they detect actions such as elderly individuals attempting to click on links in emails or access suspicious websites, and warn the elderly individual.

[1310] "Example of form 3"

[1311] Furthermore, there are mechanisms to provide warnings and advice regarding detected suspicious links and information. These mechanisms provide visual or auditory notifications to older adults and offer specific advice to protect themselves from fraud. Specifically, they may instruct older adults not to click on suspicious links or advise them to avoid accessing suspicious websites. These notifications and advice are provided in a way that is easy for older adults to understand and help them take specific actions to protect themselves from fraud.

[1312] The following describes the processing flow for each example of the form.

[1313] "Example of form 1"

[1314] Step 1: Input the content of suspicious phone calls or emails received by the elderly person into a generative AI.

[1315] Step 2: The generative AI uses natural language processing techniques to analyze the content of phone calls and emails. This analysis includes the analysis of keywords, phrases, and context.

[1316] Step 3: The generative AI determines the likelihood of fraud based on the analysis results. If it determines that there is a high probability of fraud, it warns the elderly about the information.

[1317] "Example of form 2"

[1318] Step 1: Monitor the actions taken by elderly individuals when they use the internet.

[1319] Step 2: Detect any attempts to provide suspicious links or information.

[1320] Step 3: Issue a warning to the elderly person regarding the detected operation.

[1321] "Example of form 3"

[1322] Step 1: If suspicious links or information are detected, provide visual or auditory notifications to the elderly.

[1323] Step 2: The notification will include specific advice on how to protect yourself from fraud.

[1324] Step 3: Provide notifications and advice in a way that is easy for seniors to understand, and help them take concrete actions to protect themselves from fraud.

[1325] (Example 1)

[1326] Next, we will describe Embodiment 1 of Example Form 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1327] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to provide appropriate warnings. However, conventional methods have insufficient information processing capabilities for determining the potential for fraud, resulting in a high risk of elderly people becoming victims of fraud.

[1328] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1329] In this invention, the server includes information processing means, decision means, and identification means. This makes it possible to analyze suspicious communications received by the elderly, assess the possibility of fraud, and provide appropriate warnings.

[1330] "Information processing means" refers to a device or system that has the function of analyzing suspicious communications received by elderly people and extracting important keywords using natural language processing technology.

[1331] A "decision-making tool" is a device or system that has the function of evaluating the possibility of fraud based on the analyzed communication content and generating a warning message based on the result.

[1332] "Identification means" refers to a device or system that has the function of detecting suspicious information provided by an elderly person and identifying whether that information may be related to fraud.

[1333] A "notification means" is a device or system that has the function of issuing a warning about detected suspicious information and providing warning messages to elderly people in a visual or auditory form.

[1334] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and process human language, and is used to extract meaning from text data.

[1335] A "warning message" is a message generated to alert older adults to potentially fraudulent communications, and is provided visually or audibly.

[1336] This invention is a system for analyzing suspicious communications received by elderly people and evaluating the possibility of fraud. The system includes information processing means, judgment means, identification means, and notification means.

[1337] 1. Program generation

[1338] The user generates a generative AI program using natural language processing technology. This program has the function of analyzing suspicious communications received by elderly people and determining the possibility of fraud.

[1339] 2. Program Processing

[1340] The server uses Python's natural language processing libraries, NLTK and spaCy, to analyze the communication content. The server normalizes the text data and extracts important keywords and phrases. Furthermore, it uses a generative AI model to analyze the context and assess the likelihood of fraud.

[1341] 3. Specific examples

[1342] For example, if an elderly person receives an email saying, "Your bank account has been used fraudulently. Please contact us immediately," the server will detect keywords such as "bank account," "fraudulent use," and "urgent," and determine that it is highly likely to be a scam. Based on this information, the server will generate a warning message and notify the user via their device, "This email may be a scam. Please be careful."

[1343] 4. Example of a prompt statement

[1344] An example of a prompt message for a generative AI model is, "Please analyze the content of this email and assess its likelihood of being a scam."

[1345] In this way, the system can respond quickly and accurately to suspicious communications received by the elderly, thereby preventing them from becoming victims of fraud.

[1346] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1347] Step 1:

[1348] The user inputs the content of suspicious communications received by the elderly person into the terminal. The terminal then sends this input data to the server. The input data is in text format.

[1349] Step 2:

[1350] The server preprocesses the received text data. Specifically, it uses Python's natural language processing libraries, NLTK and spaCy, to normalize the text. This includes removing unnecessary symbols and spaces, and converting words to lowercase. As a result of the preprocessing, text data in a format suitable for analysis is output.

[1351] Step 3:

[1352] The server extracts important keywords and phrases from pre-processed text data. This includes using natural language processing techniques to identify words that may be related to fraud. For example, keywords such as "bank account," "fraudulent use," and "urgent" are extracted. The extracted keywords become input data for evaluating the likelihood of fraud.

[1353] Step 4:

[1354] The server uses a generative AI model to analyze the context based on extracted keywords. The AI ​​model understands the meaning of the entire text and assesses the likelihood of fraud. This assessment determines whether the text is likely to be fraudulent. The evaluation results become input data for generating a warning message.

[1355] Step 5:

[1356] The server generates a warning message based on the evaluation results. If it is determined that there is a high probability of fraud, a warning message is created. For example, a message such as "This email may be a scam. Please be careful." is generated. The generated message is sent to the terminal.

[1357] Step 6:

[1358] The device displays warning messages received from the server to the user. Users can receive warnings visually or audibly. This helps older adults to be alerted to potentially fraudulent communications.

[1359] (Application Example 1)

[1360] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1361] There is a need to quickly and accurately determine the potential for fraud in suspicious communications received by the elderly and to issue appropriate warnings. However, conventional methods can lead to delays or incorrect judgments, putting the elderly at risk of becoming victims. To solve this problem, technology is needed that can analyze communication content in real time and determine the potential for fraud with high accuracy.

[1362] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1363] In this invention, the server includes information processing means, evaluation means, and identification means. This makes it possible to analyze received communications in real time and quickly display a warning if there is a possibility of fraud.

[1364] "Information processing means" refers to a device or program that analyzes the content of received communications and performs data processing to determine the possibility of fraud.

[1365] An "evaluation tool" is a device or program that scores the likelihood of fraud based on the analyzed communication content and makes a judgment based on the result.

[1366] "Identification means" refers to a device or program that detects suspicious elements from information provided by elderly individuals and issues a warning based on that detection.

[1367] "Notification means" refers to a device or program that provides a visual or auditory warning to a user in the event of potential fraud.

[1368] "Means of real-time analysis" refers to a device or program that instantly analyzes the content of a communication as soon as it is received and outputs the results quickly.

[1369] The system for implementing this invention involves a server and a terminal working in cooperation. The server is equipped with information processing means, evaluation means, and identification means, and analyzes the communication content received from the terminal. Specifically, the server uses a natural language processing library (e.g., spaCy, NLTK) to tokenize the communication content and extract keywords and phrases. Next, it uses a generative AI model (e.g., OpenAI GPT) to analyze the context of the extracted keywords and phrases and score the likelihood of fraud.

[1370] The device receives analysis results from the server and, if it detects a high probability of fraud, provides the user with a visual or auditory warning. This allows the user to respond quickly to suspicious communications.

[1371] As a concrete example, if a user receives an email with the subject line "Important Notice from the Bank," the device sends its contents to the server. The server extracts keywords such as "bank," "important," and "notice," and analyzes them using a generative AI model. If it is determined that the email is highly likely to be fraudulent, the device displays a warning message saying, "This email may be fraudulent. Do not click on the link."

[1372] An example of a prompt message is: "Please analyze the following email content to determine if it is a scam. Email content: 'Important notice from your bank. Please update your account information.'"

[1373] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1374] Step 1:

[1375] The terminal retrieves the content of communications (phone calls and emails) received by the user. This content is sent to the server as text data. The input is the content of communications received by the user, and the output is the transmission of text data to the server.

[1376] Step 2:

[1377] The server tokenizes the received text data using a natural language processing library (e.g., spaCy, NLTK). This process extracts keywords and phrases from the text data. The input is the text data sent from the terminal, and the output is the extracted keywords and phrases.

[1378] Step 3:

[1379] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the context of the extracted keywords and phrases. This analysis scores the likelihood of fraud. The input is the extracted keywords and phrases, and the output is a score indicating the likelihood of fraud.

[1380] Step 4:

[1381] The server determines whether there is a high probability of fraud based on the scoring results. If it determines that there is a high probability of fraud, it generates a warning message. The input is the score indicating the likelihood of fraud, and the output is the warning message.

[1382] Step 5:

[1383] The terminal notifies the user of warning messages received from the server. The notification is either visual or auditory, warning the user of potential fraud. The input is the warning message from the server, and the output is the warning notification to the user.

[1384] (Example 2)

[1385] Next, we will describe Example 2 of the morphological example. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1386] When elderly people use the internet, there is a risk of accessing suspicious links or information. Such risks can lead to phishing scams and malware infections, posing a threat to the safety of the elderly. Traditional methods are insufficient to prevent these risks, making it difficult for the elderly to use the internet with peace of mind.

[1387] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1388] In this invention, the server includes information processing means, monitoring means, and notification means. This makes it possible to immediately issue a warning when an elderly person attempts to access suspicious links or information, thereby preventing harm.

[1389] "Information processing means" refers to technology for analyzing suspicious communications received by elderly people and understanding their content.

[1390] "Monitoring measures" refer to technologies that monitor the actions of elderly people in real time and detect suspicious links or information.

[1391] "Notification means" refers to technology that generates warnings for detected suspicious links or information and informs elderly people.

[1392] A "monitoring device" is a device used to monitor the actions of elderly people in real time and has the function of recording the user's behavior.

[1393] "Analysis methods" refer to technologies used to compare suspicious links and information with a database and evaluate their safety.

[1394] "Display means" refers to technologies for displaying warning messages to the elderly visually or audibly.

[1395] This invention is a system for reducing the risk of elderly people accessing suspicious links or information when using the internet. The system includes information processing means, monitoring means, and notification means.

[1396] The server uses information processing tools to analyze suspicious communications received by elderly individuals. This analysis utilizes natural language processing technology to understand the content of the communications. Specifically, the server uses software such as a "natural language processing engine" to analyze the content of emails and messages.

[1397] The terminal uses monitoring devices to monitor the elderly person's actions in real time. Monitoring software installed on the terminal captures links the user attempts to click and URLs of websites they attempt to access, and sends them to a server. Security software is used for this monitoring.

[1398] The server uses analysis tools to compare URLs sent from the terminal against a database of known suspicious links and phishing sites. For example, it uses a "security evaluation API" to assess the security of a URL.

[1399] If suspicious links or information are detected, the server generates a warning message using notification mechanisms. This warning message is sent to the terminal and notified to the elderly visually or audibly through display mechanisms. For example, a message such as "This link is suspicious. Please stop accessing it." might be displayed.

[1400] As a concrete example, consider a scenario where an elderly person opens an email with the subject line "Receive a free gift" and attempts to click on a link within it. The device captures the link and sends it to the server. The server analyzes the link and, if it determines it may be a phishing attempt, generates a warning message saying, "This link is suspicious. Do not access it," and sends it to the device. The device displays this warning as a pop-up to the user, allowing the user to refrain from clicking the link.

[1401] An example of a prompt to input into the generation AI model would be something like, "Determine whether the link an elderly person is about to click is safe, and generate a warning message if necessary."

[1402] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1403] Step 1:

[1404] Users use the internet.

[1405] Users open emails or browse websites. For example, a user might try to click a link in an email. This action constitutes input.

[1406] Step 2:

[1407] The device monitors the user's actions.

[1408] The terminal uses monitoring software to monitor user actions in real time. Specifically, it captures the URLs of links the user attempts to click and websites they attempt to access. These URLs are then sent to the server as output.

[1409] Step 3:

[1410] The server detects suspicious links and information.

[1411] The server receives the URL sent from the terminal and uses analysis tools to compare it against a database of known suspicious links and phishing sites. This comparison with the database is data processing, and the server evaluates the safety of the URL. The result of this evaluation is the output.

[1412] Step 4:

[1413] The server generates a warning message.

[1414] Based on the analysis results, the server generates a warning message if it detects suspicious links or information. For example, it might create a message such as, "This link is suspicious. Please stop accessing it." This message is then sent to the terminal as output.

[1415] Step 5:

[1416] The device displays a warning to the user.

[1417] The device receives warning messages sent from the server and displays them to the user. Specifically, these are displayed as pop-up notifications or on-screen alerts. This display is the output, alerting the user to the warning. The user can then review the warning and choose a safe course of action.

[1418] (Application Example 2)

[1419] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal". ...

Claims

1. Equipped with a processor, The aforementioned processor, To analyze suspicious communications received by elderly individuals, a generative AI model is used to analyze the content of the communications and generate prompts to instruct the model to assess the likelihood of fraud. The generated prompt is input into the generation AI model to evaluate the likelihood of fraud, To detect suspicious information provided by elderly individuals, a generative AI model is used to analyze the information and generate prompts to instruct the system to determine whether the information is related to fraud. system.

2. The system according to claim 1, wherein the processor generates prompts to instruct the generation AI model to analyze the communication content and extract important keywords, and inputs the prompts into the generation AI model to perform the analysis.

3. The system according to claim 1, wherein the processor generates a prompt to instruct the system to provide a warning message generated by a generation AI model to an elderly person visually or audibly, and generates a warning message using the prompt.

Images