System

Abstract

A system is provided.SOLUTION: Means for receiving a call and querying caller information in a database, means for automatically blocking the call if the caller is a suspicious phone number, means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number, means for generating an automatic response message and sending it to the caller if it is determined that the risk of fraud is high, means for storing and managing the phone numbers corresponding to the blacklist in the database, means for notifying the user and his / her family in real time if a fraudulent call is detected, means for learning a newly detected fraudulent pattern; Means for improving accuracy of the analytical model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] The problem of elderly people becoming victims of special frauds is increasing. In particular, due to factors such as impaired judgment and mild dementia, elderly people are easily deceived by the sophisticated tactics of fraudsters. In addition, there are situations where family members who live far away cannot always provide support, so there is a need for a means to quickly and effectively prevent fraud. Against this background, it is necessary to develop an efficient fraud prevention system that will provide an environment in which elderly people can live safely. [Means for solving the problem]

[0005] The present invention provides an AI telephone handling service system to help elderly people avoid falling victim to special frauds. Specifically, the system includes a means for receiving a call and checking caller information against a database, a means for automatically blocking the call if the caller is a suspicious phone number, a means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number, a means for generating and sending an automatic response message to the caller if the risk of fraud is determined to be high, a means for saving and managing phone numbers that fall under a blacklist in a database, a means for notifying the user and their family in real time if a call that poses a risk of fraud is detected, and a means for learning newly detected fraud patterns and improving the accuracy of the analysis model.

[0006] A "call" is a voice communication conducted over a telephone.

[0007] "Caller information" refers to identification information such as the telephone number and name of the person you are calling.

[0008] A "database" is an electronic system that systematically organizes and stores information and allows it to be searched and updated as needed.

[0009] A "suspicious phone number" is a phone number that has been determined to be highly likely to be associated with fraud or other illegal activity.

[0010] A "blacklist" is a list of phone numbers that are deemed suspicious or unreliable.

[0011] A "whitelist" is a list of phone numbers that are deemed to be highly reliable.

[0012] A "generative AI system" is a system that uses machine learning and natural language processing technology to analyze call content and generate automatic responses.

[0013] "Analysis" refers to the act of analyzing the content of a call to understand and classify its meaning and intent.

[0014] "Risk of fraud" refers to the degree to which the content of a call is likely to be fraudulent.

[0015] "Auto-Response Message" means a system-generated automated message sent to a caller.

[0016] "Notification" refers to the act of the system informing the user or their family of a specific event or information.

[0017] A "continuous learning system" is an AI technology that continues to learn by incorporating new data and information, improving the system's performance and accuracy. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9]1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0020] First, the terms used in the following description will be explained.

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0022] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0023] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0039] The present invention is an AI telephone handling service system for helping elderly people avoid falling victim to special fraud, and can be implemented as follows.

[0040] System Overview

[0041] The system consists of the following main components:

[0042] 1. Call Reception Server

[0043] 2. Generative AI Systems

[0044] 3. Database System

[0045] 4. User Notification System

[0046] 5. Continuous Learning System

[0047] Program processing

[0048] Call acceptance server

[0049] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[0050] Generative AI System

[0051] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0052] Database System

[0053] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[0054] User Notification System

[0055] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0056] Continuous Learning System

[0057] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0058] Specific examples

[0059] Receiving and analyzing calls

[0060] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the database for the caller's phone number. If the call is not blacklisted, the server sends the call to the AI ​​system.

[0061] Analysis and automatic response by generative AI

[0062] The AI ​​system analyzes the content of calls in real time, detecting keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, the AI ​​generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[0063] Call blocking and notifications

[0064] The server plays the automated response message sent by the AI ​​system to the caller and disconnects the call. At the same time, the target phone number is added to the blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[0065] Continuous learning and updates

[0066] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0067] In this way, this system can protect the elderly from the risk of special fraud and provide an environment where they can live in peace of mind.

[0068] The processing flow will be explained below.

[0069] Step 1:

[0070] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[0071] Step 2:

[0072] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[0073] Step 3:

[0074] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[0075] Step 4:

[0076] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generation AI system.

[0077] Step 5:

[0078] The generative AI system analyzes incoming call data to assess fraud risk, detecting keywords and conversation patterns such as "identity verification," "account," and "funds."

[0079] Step 6:

[0080] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0081] Step 7:

[0082] The server plays the automated response message sent by the generating AI system to the caller and then disconnects the call.

[0083] Step 8:

[0084] The server adds the caller's phone number to a blacklist in its database, automatically blocking future calls.

[0085] Step 9:

[0086] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected via SMS, notification app, or email.

[0087] Step 10:

[0088] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the generative AI system's model, thereby improving the analysis accuracy of the entire system.

[0089] Example 1

[0090] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0091] In modern society, the risk of elderly people becoming victims of special frauds is increasing. Such frauds are becoming more sophisticated, with a particularly rapid increase in fraud methods over the phone. Elderly people have difficulty recognizing fraud and are often more susceptible to falling victim. Therefore, there is a need for a system that can automatically assess fraud risks and respond quickly. Furthermore, there is a need for a means to notify elderly people and their families of fraud risks in real time, and a system that can continuously respond to new fraud patterns is also needed.

[0092] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0093] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means for recording the content of calls and sending it to a generative AI model system for analysis; and means for generating an automated response message based on the analysis results and sending it to the server. This protects elderly people from the risk of special fraud and allows them to receive prompt fraud warnings. It also enables continuous learning and improvement of the accuracy of the analysis model to respond to new fraud techniques.

[0094] "Means for receiving calls and checking the database for caller information" refers to a function that accepts external telephone calls and checks the caller's telephone number against an internal database to see if the caller matches the registered information.

[0095] "Means for automatically blocking calls when the caller is a suspicious phone number" refers to a function that automatically disconnects the call and prevents the call from being forwarded if the caller's phone number is blacklisted as a result of a database check.

[0096] "A means of analyzing the content of calls in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that uses voice recognition to instantly analyze the content of calls and determine whether there is a possibility of fraud if the call is not on the blacklist.

[0097] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" refers to a function that automatically generates a standard response message and sends it to the caller when the analysis of the content of the call indicates a high risk of fraud.

[0098] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for recording telephone numbers of suspicious callers in a database and continuously managing and updating that information.

[0099] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that sends an immediate warning notification to users and their families when a call that poses a high risk of fraud is detected.

[0100] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function for improving the accuracy of fraud detection by accumulating newly detected fraud methods and patterns as learning data and periodically updating the analysis algorithm model.

[0101] "Means for recording the contents of a call and transmitting it to a generative AI model system for analysis" refers to a function that records the audio during a call and transmits the recorded data to an analysis system that uses a generative AI model, thereby performing processing to analyze the contents of the call.

[0102] "Means for generating an automatic response message based on the analysis results and sending it to the server" refers to a function that creates an appropriate automatic response message based on the results of analysis by the generative AI model system and sends that message to the server.

[0103] MODE FOR CARRYING OUT THE INVENTION

[0104] This invention is an AI telephone handling service system designed to protect elderly people from falling victim to special fraud. This system consists of the following main components: a call reception server, a generation AI system, a database system, a user notification system, and a continuous learning system. Below, we provide a detailed explanation of each component and explain how the system works.

[0105] System Overview

[0106] Call acceptance server

[0107] The call reception server monitors all incoming calls. When a new call comes in, it captures the call information and checks whether it is an unknown phone number. The server then queries the caller's phone number in a database system to determine whether the number is blacklisted. If the caller is not blacklisted, it records the call and sends the data to the generation AI system.

[0108] Examples of hardware and software used: VoIP devices, communication control servers

[0109] Generative AI System

[0110] The generative AI system analyzes incoming call data and assesses the risk of fraud. The system includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an automated response message and sends it to the server.

[0111] Examples of hardware and software used: Natural Language Processing (NLP) engine, Python, TENSORFLOW (registered trademark) library

[0112] Database System

[0113] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the AI ​​system determines to be suspicious to a blacklist in the database, automatically blocking future calls.

[0114] Examples of hardware and software used: SQL database server

[0115] User Notification System

[0116] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0117] Examples of hardware and software used: notification server, SMS gateway, notification application

[0118] Continuous Learning System

[0119] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0120] Examples of hardware and software used: machine learning platform, data storage system

[0121] Specific examples

[0122] Receiving and analyzing calls

[0123] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to the AI ​​system.

[0124] Analysis and automatic response by generative AI

[0125] The AI ​​generation system analyzes the content of calls in real time. For example, it detects keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, it generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[0126] Call blocking and notifications

[0127] The server plays the automated message sent by the AI ​​system to the caller and disconnects the call. At the same time, the server adds the phone number to a blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[0128] Continuous learning and updates

[0129] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0130] Prompt Sentence Examples

[0131] "Analyze the content of this call, and if suspicious keywords (such as 'identity verification,' 'account,' 'funds,' etc.) are detected, determine that it may be a scam and generate an automated response message."

[0132] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0133] Step 1:

[0134] The server monitors all incoming calls. When a new call is received, it captures the call information (caller's phone number, date and time, etc.). (Input): New incoming call. (Data processing): Capture and record call information. (Output): Caller information.

[0135] Step 2:

[0136] The server queries the caller's phone number in a database system. (Input): Caller information for incoming calls. (Data calculation): Query the database. (Output): Query results (blacklist match).

[0137] Step 3:

[0138] The server judges the query result and checks whether the caller is on the blacklist. If the caller is on the blacklist, the server will automatically block the call. (Input): Database query result. (Data calculation): Judgment process. (Output): Call blocking.

[0139] Step 4:

[0140] If the caller is not on the blacklist, the server records the call and sends the data to the generation AI system. (Input): Call information that does not fall under the blacklist. (Data processing): Recording of call content and data transmission. (Output): Recorded data.

[0141] Step 5:

[0142] The generative AI system analyzes the received call data and assesses the risk of fraud. (Input): Recorded data. (Data calculation): Analysis of call content and risk assessment using natural language processing. (Output): Risk assessment results (presence or absence of fraud risk and its degree).

[0143] Step 6:

[0144] If the risk of fraud is determined to be high, the AI ​​system generates an automatic response message and sends it to the server. (Input): Risk assessment result. (Data calculation): Generation of automatic response message. (Output): Automatic response message.

[0145] Step 7:

[0146] The server plays the automated response message sent by the AI ​​generation system to the caller and disconnects the call. At the same time, it adds the target phone number to the blacklist. (Input): Auto-response message, caller information. (Data processing): Plays the message and disconnects the call, updates the blacklist. (Output): Phone number added to the blacklist.

[0147] Step 8:

[0148] The user notification system sends real-time notifications to users and their families when fraudulent calls are detected. (Input): Risk assessment results and blacklist update information. (Data calculation): Notification message generation. (Output): Notification message (SMS, notification app, email).

[0149] Step 9:

[0150] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the model of the generative AI system. (Input): Newly detected fraud data. (Data calculation): Update training data and retrain the model. (Output): Improved AI model.

[0151] (Application example 1)

[0152] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0153] With the elderly currently at a higher risk of falling victim to special frauds, there is a need for a system to protect them from telephone fraud. However, existing systems have difficulty analyzing call content in real time and immediately assessing the risk of fraud. Even if calls with a high risk of fraud are detected, appropriate measures may not be taken promptly, making them insufficient to ensure the safety of the elderly. Furthermore, there is a risk that responses to new fraud methods will be delayed, reducing the accuracy of the system. It is necessary to solve these issues and provide an environment in which the elderly can live with peace of mind.

[0154] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0155] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying users and their families in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means operated via a mobile communication terminal with a call monitoring function; and means with a security function for sending the generated automated response message to a designated recipient. This safely protects elderly people from special frauds, analyzes suspicious calls in real time to take appropriate measures, and enables elderly people and their families to quickly understand the risks.

[0156] A "mobile communication terminal with call monitoring function" refers to a portable communication device that has the function of receiving calls and monitoring their content.

[0157] The "means for querying a database for caller information" refers to a system that has the function of querying and confirming caller information from a database that stores information about the caller.

[0158] "Automatic call blocking" is a feature that automatically blocks calls when a suspicious phone number or call is detected.

[0159] "Means for assessing the risk of fraud" refers to a function that analyzes the content of a call and determines the likelihood that the call is fraudulent.

[0160] "Means for generating and sending automatic response messages to callers" refers to a function that automatically creates a pre-defined message and sends it to callers when it is determined that there is a high risk of fraud.

[0161] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function that adds telephone numbers that are determined to be at high risk of fraud to a specific list and manages that list in a database.

[0162] "Means of notifying users and their families in real time" is a function that quickly notifies users and their families of any calls that pose a risk of fraud when such calls are detected.

[0163] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function that allows the system to learn newly discovered fraud methods and patterns, update the analysis model based on them, and improve accuracy.

[0164] The present invention relates to a system for protecting elderly people from special frauds, which is implemented using a mobile communication terminal and a server. Specific embodiments of the system are described below.

[0165] 1. System Configuration

[0166] The system consists of the following main components:

[0167] 1. Mobile communication terminals

[0168] 2. Call Reception Server

[0169] 3. Generative AI Systems

[0170] 4. Database Systems

[0171] 5. User Notification System

[0172] 6. Continuous Learning System

[0173] 2. System Operation

[0174] Mobile communication terminal

[0175] A mobile communication terminal is a portable communication device equipped with a call monitoring function. When a call is received, the terminal records the call and transmits the caller information to a server.

[0176] Call acceptance server

[0177] The server receives calls and checks the caller's phone number against a database. If a call comes from an unknown number, it records the call and sends it to a generative AI system. If the number is suspicious, the call is automatically blocked.

[0178] Generative AI System

[0179] The generative AI system analyzes the content of calls in real time and assesses the risk of fraud. The AI ​​model converts the voice data into text and analyzes the text to determine the risk of fraud. If it determines that the risk of fraud is high, it generates an appropriate automated response message and sends it to the server.

[0180] Database System

[0181] The database system stores and manages information on suspicious phone numbers and fraudulent methods. For example, a blacklist is managed in this database, and the server adds new blacklist items based on information from the generative AI system.

[0182] User Notification System

[0183] The user notification system will send real-time notifications to users and their families when a fraudulent call is detected. This notification will be sent via SMS, a notification app, or email, allowing users and their families to quickly understand the risk.

[0184] Continuous Learning System

[0185] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0186] 3. Hardware and Software

[0187] The following hardware and software are used to implement this system:

[0188] Hardware: Smartphones, servers

[0189] Software: Python, Requests library, smtplib, custom API (call acceptance, generative AI, database management)

[0190] 4. Examples and prompts

[0191] Specific examples

[0192] An elderly person installs a smartphone app. One day, the elderly person receives a call from an unknown number. The app monitors the call and determines it to be suspicious. A notification is sent to the family saying, "There was a suspicious call from 0123456789."

[0193] Prompt example

[0194] "Please explain the functionality of an AI application that analyzes calls from unknown numbers in real time to protect seniors from specialized fraud."

[0195] This will ensure that elderly people are safely protected from special fraud, suspicious calls will be analyzed in real time and appropriate measures will be taken, and elderly people and their families will be able to quickly understand the risks.

[0196] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0197] Step 1:

[0198] The server receives calls from mobile communication terminals, obtains the caller's phone number, and checks the caller information against a database to see if the phone number is suspicious.

[0199] Specifically, it receives the caller's phone number as input, searches for the corresponding phone number in the database, and returns a flag indicating whether the caller's phone number is suspicious.

[0200] Step 2:

[0201] If the server determines that the caller is a suspicious number, it will automatically block the call.

[0202] Specifically, it receives a suspicious phone number flag as input, blocks the call, and returns a confirmation message that the call has been blocked as output.

[0203] Step 3:

[0204] If the caller is not a suspicious phone number, the server records the call and sends the data to the generation AI system.

[0205] Specifically, it receives voice call data as input, sends it to the AI ​​analysis system, and returns a message confirming the transmission of the voice call data as output.

[0206] Step 4:

[0207] The generative AI system analyzes incoming call content in real time and assesses the risk of fraud.

[0208] Specifically, the system receives voice call data as input, performs text analysis using an AI model, and returns the fraud risk assessment results to the server as output.

[0209] Step 5:

[0210] The server receives the evaluation results from the generative AI system and, if it determines that there is a high risk of fraud, generates an automated response message to send to the caller.

[0211] Specifically, it receives the fraud risk assessment result as input, generates an automatic response message, and returns the response message to be sent to the caller as output.

[0212] Step 6:

[0213] The server adds phone numbers that are deemed to pose a fraud risk to a blacklist in its database, automatically blocking future calls.

[0214] Specifically, it receives a fraudulent phone number as input, registers it in a blacklist in the database, and returns a message confirming the registration to the blacklist as output.

[0215] Step 7:

[0216] The user notification system notifies users and their families in real time when calls that pose a risk of fraud are detected.

[0217] It takes the fraud risk assessment result as input, generates notification data, and sends notifications via SMS, notification app, or email as output.

[0218] Step 8:

[0219] The continuous learning system stores newly detected fraud patterns as training data and periodically updates the generative AI system's model.

[0220] Specifically, it receives new fraud pattern data as input, stores it in the training database, and returns a confirmation message for the updated AI model as output.

[0221] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0222] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds, and in particular, improves the accuracy of fraud detection by combining it with an emotion engine that recognizes the user's emotions. The following describes specific embodiments of the present invention.

[0223] System Overview

[0224] The system consists of the following main components:

[0225] 1. Call Reception Server

[0226] 2. Generative AI Systems

[0227] 3. Database System

[0228] 4. User Notification System

[0229] 5. Continuous Learning System

[0230] 6. Emotion Engine

[0231] Program processing

[0232] Call acceptance server

[0233] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[0234] Generative AI System

[0235] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0236] Database System

[0237] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[0238] User Notification System

[0239] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0240] Continuous Learning System

[0241] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0242] Emotion Engine

[0243] The emotion engine analyzes the user's emotions during the call and sends the results to the generative AI system. The emotion engine analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc.

[0244] Specific examples

[0245] Receiving and analyzing calls

[0246] For example, if a call comes in to a senior citizen's phone from an unknown number, the server will record the call and query the caller's phone number against a database. If the database does not match any blacklist, the call will be sent to the generative AI system and emotion engine.

[0247] Collaboration between analysis and generative AI using an emotion engine

[0248] The generative AI system analyzes the call content in real time, detecting keywords such as "identity verification," "account," and "funds." At the same time, the emotion engine analyzes the voice and evaluates the user's emotional state. If the emotion engine determines that the user's emotions are abnormal, such as anxiety or fear, the generative AI will take this into account when assessing the fraud risk.

[0249] Auto-response and blacklist management

[0250] If the AI ​​system determines that there is a high risk of fraud, it generates an automated response message such as "We are currently unavailable. We will contact you shortly" and sends it to the server. The server plays this message to the caller and disconnects the call. The server then adds the caller's phone number to a blacklist in its database.

[0251] Notification and follow-up

[0252] Based on the analysis results of the emotion engine and the risk assessment of the generative AI, the user notification system will send warning messages such as "A suspicious call has been detected" to the elderly person's family members, and will also provide further detailed advice and follow-up notifications if necessary.

[0253] Continuous learning and updates

[0254] The continuous learning system saves the detected fraud patterns and data from the emotion engine as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving the fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0255] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[0256] The processing flow will be explained below.

[0257] Step 1:

[0258] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[0259] Step 2:

[0260] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[0261] Step 3:

[0262] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[0263] Step 4:

[0264] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generative AI system and emotion engine.

[0265] Step 5:

[0266] The generative AI system analyzes the transmitted call data and assesses the risk of specialized fraud. The analysis involves detecting keywords and conversation patterns, such as "identity verification," "account," and "funds."

[0267] Step 6:

[0268] The emotion engine analyzes voice data during a call to detect the user's tone and emotional fluctuations, thereby assessing whether the user is experiencing emotions such as anxiety, fear, or anger.

[0269] Step 7:

[0270] The generative AI system adjusts the fraud risk assessment based on the sentiment analysis results from the emotion engine. For example, if the emotion engine detects anxiety or fear in the user, it will determine that the fraud risk is high.

[0271] Step 8:

[0272] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0273] Step 9:

[0274] The server plays the automated response message sent by the AI ​​system to the caller, disconnects the call, and adds the caller's phone number to a blacklist in the database.

[0275] Step 10:

[0276] The user notification system will send real-time notifications to users and their families based on the analysis results of the emotion engine and the risk assessment of the generative AI system via SMS, notification app, or email.

[0277] Step 11:

[0278] The notification message will include something like, "A suspicious call has been detected. Please contact your system administrator for more information." In some cases, more detailed advice or follow-up notifications will be provided.

[0279] Step 12:

[0280] The continuous learning system saves detected fraud patterns and emotion engine data as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving fraud detection accuracy in future iterations.

[0281] Example 2

[0282] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0283] There is a problem that exposes elderly people to the risk of special fraud. Existing telephone systems may not be able to detect fraud risks, increasing the likelihood that elderly people will become victims of fraud. In addition, it is difficult to quickly notify and respond to fraud risks, so prompt measures are needed to minimize damage.

[0284] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for analyzing the user's emotions and reflecting them in the fraud risk assessment; means for generating and sending an automatic response message to the caller if the fraud risk is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call with a risk of fraud is detected; and means for learning newly detected fraud patterns and improving the accuracy of the analysis model. This protects elderly people from fraud, enables rapid and accurate detection of fraud risks, and enables appropriate responses.

[0285] "Means for receiving calls and checking the database for caller information" is a function that detects an incoming call and checks the caller's telephone number against a database to confirm it.

[0286] "Means to automatically block calls if the caller is a suspicious phone number" is a function that automatically prevents calls from being received if the caller's phone number is included in a blacklist in the database.

[0287] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of a conversation during a call in real time and determines the risk of fraud based on that content.

[0288] "Means for analyzing user emotions and reflecting them in fraud risk assessment" is a function that analyzes changes in the user's tone of voice and emotions during a call and uses the results to help assess fraud risk.

[0289] "Means for generating and sending an automatic response message to a caller when a high risk of fraud is determined" is a function that sends an automatically created message to a caller for calls that are assessed as having a high risk of fraud.

[0290] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for registering telephone numbers that are determined to pose a risk of fraud in a database and for retaining and managing that information.

[0291] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that notifies users and their families in real time when a call that poses a high risk of fraud occurs.

[0292] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that learns newly discovered fraud methods and patterns and improves the accuracy of the fraud detection model for the entire system.

[0293] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds. In particular, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of fraud detection is improved. This system is realized using hardware and software. The following describes in detail an embodiment of the present invention.

[0294] System Overview

[0295] The system consists of the following main components:

[0296] 1. Call Reception Server

[0297] 2. Generative AI Systems

[0298] 3. Database System

[0299] 4. User Notification System

[0300] 5. Continuous Learning System

[0301] 6. Emotion Engine

[0302] Call acceptance server

[0303] The server monitors all incoming calls, and if a call comes from an unknown phone number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the generation AI system. The hardware used is a standard server and telephone system for call management. The software used is call recording software and database query software.

[0304] Examples:

[0305] When an elderly user receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to a generative AI system and emotion engine.

[0306] Generative AI System

[0307] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server. A general language model is used as the generative AI model.

[0308] Examples:

[0309] The generative AI system analyzes call content in real time, detecting keywords such as "identity verification," "account," and "funds," while also taking into account the analysis results of the emotion engine to assess fraud risk.

[0310] Example prompt sentence:

[0311] "Assess your risk of fraud based on the following conversation: 'Hello, this is from your bank and I need to verify your account.'"

[0312] Database System

[0313] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the generative AI system deems suspicious to a blacklist in the database, automatically blocking future calls. The hardware used is a standard database server, and the software is a common database management system.

[0314] Examples:

[0315] The server adds the phone number of callers it deems a high risk of fraud to a database and automatically blocks future calls.

[0316] User Notification System

[0317] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. The hardware used is a communication server for sending notifications, and the software used is a notification app or SMS sending system.

[0318] Examples:

[0319] Based on the analysis results of the emotion engine and the risk assessment by the generating AI, a warning message such as "A suspicious call has been detected" is sent via SMS to the elderly person's family.

[0320] Continuous Learning System

[0321] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0322] Examples:

[0323] Newly detected fraud patterns, along with data from the emotion engine, are saved as training data to update the model of the generative AI system.

[0324] Emotion Engine

[0325] The emotion engine analyzes the user's emotions during a call and sends the results to the generative AI system. It analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc. The hardware uses a computer for voice analysis, and the software uses an emotion analysis algorithm.

[0326] Examples:

[0327] The call content is analyzed, and if the user's emotions indicate anxiety or fear, the results are sent to a generative AI system, which takes this into account when assessing fraud risk.

[0328] Example prompt sentence:

[0329] "Analyze the emotions from this audio data and tell us what emotions the user is feeling."

[0330] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[0331] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0332] Step 1:

[0333] The server detects incoming phone calls and monitors them. If an incoming call comes from an unknown phone number, the server records the call and records details such as the call timestamp, the caller's phone number, and the call length. Specifically, when an elderly user receives a call from an unknown number on their device, the server begins recording the call. The input is the call details, and the output is the recorded audio data.

[0334] Step 2:

[0335] The server queries the caller information of the recorded call in a database system to see if the number is on the blacklist. Specifically, the server queries the caller's phone number in the database and gets the result "not on the blacklist." The input is the caller's phone number, and the output is the database query result.

[0336] Step 3:

[0337] If the server confirms that the caller is not on the blacklist, it records the call and sends the data to the generative AI system and emotion engine. Specifically, after the call is recorded, the server sends the audio data to the generative AI system and emotion engine. The input is the audio data of the recorded call, and the output is the transfer of data to the generative AI system and emotion engine.

[0338] Step 4:

[0339] The generative AI system analyzes the received voice data and detects specific keywords and patterns. The emotion engine also analyzes the voice data and evaluates the user's emotional state. Specifically, the generative AI system detects keywords such as "identity verification," "account," and "funds" from the call, and the emotion engine analyzes the tone and speed of the user's voice to determine whether they are feeling anxious. The input is the voice data of the call content, and the output is specific keywords and the user's emotional evaluation result.

[0340] Step 5:

[0341] The server evaluates whether the risk of fraud is high based on the fraud risk assessment results provided by the generative AI system and the analysis results of the emotion engine. Specifically, if the generative AI system returns a high risk score and the emotion engine concludes that the user is anxious, the server determines that the risk of fraud is high. The inputs are the assessment results of the generative AI system and the analysis results of the emotion engine, and the output is the risk assessment result.

[0342] Step 6:

[0343] If the server determines that there is a high risk of fraud, it sends the generated auto-answer message to the caller and disconnects the call.Specifically, a message such as "There is no one in charge at the moment. We will contact you later" is played and the call is disconnected.The input is a template for the auto-answer message, and the output is sending a message to the caller and disconnecting the call.

[0344] Step 7:

[0345] The server adds the caller's phone number to a blacklist in the database system. Specifically, the server registers the caller's phone number in the database, and future calls are automatically blocked. The input is the caller's phone number, and the output is the registration result in the database.

[0346] Step 8:

[0347] The user notification system notifies users and their families in real time about detected fraud risks. Specifically, it sends a warning message to the elderly family member's smartphone via SMS saying, "A suspicious call has been detected." The input is the fraud risk assessment result, and the output is the sending of a notification message.

[0348] Step 9:

[0349] The continuous learning system saves newly detected fraud patterns and emotion data as learning data and periodically updates the generative AI system and emotion engine models. Specifically, the detected fraud patterns and emotion data are saved in a database, and the model is periodically updated. The input is the newly detected fraud patterns and emotion data, and the output is the updated AI model.

[0350] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[0351] (Application example 2)

[0352] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0353] In recent years, the number of elderly people who fall victim to special frauds over the phone has been increasing. However, conventional fraud prevention systems are finding it difficult to respond effectively to the increasing diversity and sophistication of fraud methods. In addition, since it is difficult for elderly people to judge the content of phone calls themselves, there is a need for a system that can prevent fraud damage before it occurs. In particular, the development of an advanced fraud prevention system that automatically responds to fraud methods and combines real-time emotion analysis is a challenge.

[0354] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automatic response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; emotion analysis means for evaluating the user's emotional state; and means for complementing the fraud risk assessment based on the user's emotional state. This significantly reduces the risk of elderly people falling victim to special telephone frauds and provides an environment in which they can live safely.

[0355] "Means for receiving calls and checking the caller information against a database" is a function that automatically obtains information about an incoming call and compares the caller information with a pre-registered database.

[0356] "Means for automatically blocking calls when the caller is a suspicious phone number" is a function that automatically blocks incoming calls from suspicious phone numbers based on the results of a database match.

[0357] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of the conversation in real time during a call and assesses the possibility of fraud based on that content.

[0358] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" is a function that automatically generates a response message and sends it to the caller when the system determines that there is a high risk of fraud.

[0359] "Means for storing and managing blacklisted telephone numbers in a database" refers to a management function that stores telephone numbers of callers deemed suspicious in a database and consistently blocks subsequent calls.

[0360] "Means of notifying users and their families in real time when calls that pose a high risk of fraud are detected" is a function that immediately sends a warning to users and their families when calls that pose a high risk of fraud are detected.

[0361] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that adds newly discovered fraud methods and patterns as learning data and improves the accuracy of the analysis model.

[0362] The "emotion analysis means for evaluating the user's emotional state" is a function that analyzes the user's tone of voice and choice of words during a call and evaluates their emotional state.

[0363] "Means to complement fraud risk assessment based on emotional state" is a function that further refines fraud risk assessment based on the results of user emotional analysis.

[0364] To implement the present invention, a system including the following procedures and components is used.

[0365] System Overview

[0366] The system consists of a server that receives and analyzes calls in real time, a terminal that monitors the user's call status, and a notification system that notifies users and their families based on the results of the call analysis.

[0367] 1. Receiving calls and querying caller information

[0368] The server automatically records calls received by the user and checks the caller information against a database that includes a blacklist of suspicious phone numbers.

[0369] 2. Real-time analysis and automatic response

[0370] The server analyzes the call content in real time and assesses the risk of fraud based on the content of the conversation and the keywords used.

[0371] If the risk of fraud is determined to be high, the server generates an automated response message and sends it to the caller, designed to prevent fraud before it occurs.

[0372] 3. Emotion analysis

[0373] The server uses an emotion engine to analyze the tone, rate and emotional fluctuations of the user's voice during the call to assess the user's emotional state.

[0374] Sentiment analysis results are used to complement fraud risk assessment.

[0375] 4. Notification System

[0376] If a fraudulent call is detected, the notification system will send a real-time notification to the user and their family via SMS, notification app, or email.

[0377] 5. Blacklist Management

[0378] The server stores phone numbers that are deemed suspicious in a blacklist in its database and automatically blocks future calls.

[0379] 6. Continuous Learning System

[0380] The continuous learning system saves newly detected fraud patterns as training data to improve the accuracy of the analysis model.

[0381] Hardware and software used

[0382] 1. Hardware

[0383] Server: Data processing device for analysis and notification

[0384] Smartphone: Records user calls and sends them to a server

[0385] 2. Software

[0386] Call reception app: Records user calls and sends the data to a server

[0387] Generative AI system: Analyzes call content and assesses fraud risk

[0388] Emotion Engine: Analyzes the user's emotional state

[0389] Notification app: Sends notifications about fraud risks

[0390] Specific examples

[0391] 1. Receiving and analyzing calls

[0392] When a user receives a call from an unknown number, the server records the call and checks the caller's number against a database. If the call is not blacklisted, the server analyzes the call in real time.

[0393] 2. Sentiment Analysis and Risk Assessment

[0394] The server uses an emotion engine to analyze the user's emotional state, and the generative AI system analyzes keywords in the call content. The results of the emotion analysis and the call content analysis are combined to assess the fraud risk.

[0395] 3. Auto-replies and notifications

[0396] If the risk of fraud is deemed high, the server generates an automated response message and sends it to the caller, while simultaneously sending real-time notifications to the user and their family members via SMS, notification app, or email.

[0397] Prompt Sentence Examples

[0398] Call content: {Call text}

[0399] Emotional state: {User emotion}

[0400] Criteria: Assess the fraud risk and return a "High Fraud Risk" message if the risk is high.

[0401] In this way, this system reduces the risk of users becoming victims of special fraud and provides an environment in which users can make calls with peace of mind.

[0402] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0403] Step 1:

[0404] When a user receives a call, the device records the call and obtains the caller's phone number. The obtained call data and caller information are sent to the server. The input is the incoming call's voice data and the caller's phone number, and the output is that the data is sent to the server.

[0405] Step 2:

[0406] The server checks the received caller information against the database. If the check finds that the caller is registered on the blacklist, the server automatically blocks the call. The input is the caller's phone number, and the output is the check result and the call blocking process.

[0407] Step 3:

[0408] If the caller is not on the blacklist, the server analyzes the call in real time using a generative AI system that performs keyword analysis and pattern matching on the call content. The input is the audio data of the call, and the output is the fraud risk assessment result.

[0409] Step 4:

[0410] The server uses an emotion engine to analyze the tone, speed, and emotional fluctuations of the user's voice during the call to evaluate the user's emotional state. The input is the voice data of the call, and the output is the evaluation result of the user's emotional state.

[0411] Step 5:

[0412] The server evaluates the risk of fraud by combining the results of keyword analysis by the generative AI system and the results of the emotional state evaluation by the emotion engine. If the risk of fraud is determined to be high, the server generates an automatic response message and sends it to the caller. The input is the analysis results of the generative AI system and the evaluation results of the emotion engine, and the output is the automatic response message.

[0413] Step 6:

[0414] If a fraud risk is detected, the server sends a notification to the user and their family members via SMS, a notification app, or email. The input is the fraud risk assessment result, and the output is the notification message.

[0415] Step 7:

[0416] The server adds any phone numbers it deems suspicious to a blacklist in its database, automatically blocking future calls. The input is the suspicious phone number, and the output is the registration status in the database.

[0417] Step 8:

[0418] The continuous learning system improves accuracy by saving newly detected fraud patterns as training data and updating the analytical models of the generative AI system and emotion engine. The input is the newly detected fraud patterns, and the output is the updated models.

[0419] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0420] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0421] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0422] [Second embodiment]

[0423] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0424] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0425] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0426] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0427] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0428] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0429] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0430] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0431] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0432] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0433] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0434] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0435] The present invention is an AI telephone handling service system for helping elderly people avoid falling victim to special fraud, and can be implemented as follows.

[0436] System Overview

[0437] The system consists of the following main components:

[0438] 1. Call Reception Server

[0439] 2. Generative AI Systems

[0440] 3. Database System

[0441] 4. User Notification System

[0442] 5. Continuous Learning System

[0443] Program processing

[0444] Call acceptance server

[0445] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[0446] Generative AI System

[0447] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0448] Database System

[0449] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[0450] User Notification System

[0451] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0452] Continuous Learning System

[0453] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0454] Specific examples

[0455] Receiving and analyzing calls

[0456] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the database for the caller's phone number. If the call is not blacklisted, the server sends the call to the AI ​​system.

[0457] Analysis and automatic response by generative AI

[0458] The AI ​​system analyzes the content of calls in real time, detecting keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, the AI ​​generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[0459] Call blocking and notifications

[0460] The server plays the automated response message sent by the AI ​​system to the caller and disconnects the call. At the same time, the target phone number is added to the blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[0461] Continuous learning and updates

[0462] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0463] In this way, this system can protect the elderly from the risk of special fraud and provide an environment where they can live in peace of mind.

[0464] The processing flow will be explained below.

[0465] Step 1:

[0466] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[0467] Step 2:

[0468] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[0469] Step 3:

[0470] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[0471] Step 4:

[0472] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generation AI system.

[0473] Step 5:

[0474] The generative AI system analyzes incoming call data to assess fraud risk, detecting keywords and conversation patterns such as "identity verification," "account," and "funds."

[0475] Step 6:

[0476] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0477] Step 7:

[0478] The server plays the automated response message sent by the generating AI system to the caller and then disconnects the call.

[0479] Step 8:

[0480] The server adds the caller's phone number to a blacklist in its database, automatically blocking future calls.

[0481] Step 9:

[0482] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected via SMS, notification app, or email.

[0483] Step 10:

[0484] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the generative AI system's model, thereby improving the analysis accuracy of the entire system.

[0485] Example 1

[0486] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0487] In modern society, the risk of elderly people becoming victims of special frauds is increasing. Such frauds are becoming more sophisticated, with a particularly rapid increase in fraud methods over the phone. Elderly people have difficulty recognizing fraud and are often more susceptible to falling victim. Therefore, there is a need for a system that can automatically assess fraud risks and respond quickly. Furthermore, there is a need for a means to notify elderly people and their families of fraud risks in real time, and a system that can continuously respond to new fraud patterns is also needed.

[0488] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0489] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means for recording the content of calls and sending it to a generative AI model system for analysis; and means for generating an automated response message based on the analysis results and sending it to the server. This protects elderly people from the risk of special fraud and allows them to receive prompt fraud warnings. It also enables continuous learning and improvement of the accuracy of the analysis model to respond to new fraud techniques.

[0490] "Means for receiving calls and checking the database for caller information" refers to a function that accepts external telephone calls and checks the caller's telephone number against an internal database to see if the caller matches the registered information.

[0491] "Means for automatically blocking calls when the caller is a suspicious phone number" refers to a function that automatically disconnects the call and prevents the call from being forwarded if the caller's phone number is blacklisted as a result of a database check.

[0492] "A means of analyzing the content of calls in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that uses voice recognition to instantly analyze the content of calls and determine whether there is a possibility of fraud if the call is not on the blacklist.

[0493] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" refers to a function that automatically generates a standard response message and sends it to the caller when the analysis of the content of the call indicates a high risk of fraud.

[0494] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for recording telephone numbers of suspicious callers in a database and continuously managing and updating that information.

[0495] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that sends an immediate warning notification to users and their families when a call that poses a high risk of fraud is detected.

[0496] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function for improving the accuracy of fraud detection by accumulating newly detected fraud methods and patterns as learning data and periodically updating the analysis algorithm model.

[0497] "Means for recording the contents of a call and transmitting it to a generative AI model system for analysis" refers to a function that records the audio during a call and transmits the recorded data to an analysis system that uses a generative AI model, thereby performing processing to analyze the contents of the call.

[0498] "Means for generating an automatic response message based on the analysis results and sending it to the server" refers to a function that creates an appropriate automatic response message based on the results of analysis by the generative AI model system and sends that message to the server.

[0499] MODE FOR CARRYING OUT THE INVENTION

[0500] This invention is an AI telephone handling service system designed to protect elderly people from falling victim to special fraud. This system consists of the following main components: a call reception server, a generation AI system, a database system, a user notification system, and a continuous learning system. Below, we provide a detailed explanation of each component and explain how the system works.

[0501] System Overview

[0502] Call acceptance server

[0503] The call reception server monitors all incoming calls. When a new call comes in, it captures the call information and checks whether it is an unknown phone number. The server then queries the caller's phone number in a database system to determine whether the number is blacklisted. If the caller is not blacklisted, it records the call and sends the data to the generation AI system.

[0504] Examples of hardware and software used: VoIP devices, communication control servers

[0505] Generative AI System

[0506] The generative AI system analyzes incoming call data and assesses the risk of fraud. The system includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an automated response message and sends it to the server.

[0507] Examples of hardware and software used: Natural Language Processing (NLP) engine, Python, TensorFlow library

[0508] Database System

[0509] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the AI ​​system determines to be suspicious to a blacklist in the database, automatically blocking future calls.

[0510] Examples of hardware and software used: SQL database server

[0511] User Notification System

[0512] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0513] Examples of hardware and software used: notification server, SMS gateway, notification application

[0514] Continuous Learning System

[0515] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0516] Examples of hardware and software used: machine learning platform, data storage system

[0517] Specific examples

[0518] Receiving and analyzing calls

[0519] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to the AI ​​system.

[0520] Analysis and automatic response by generative AI

[0521] The AI ​​generation system analyzes the content of calls in real time. For example, it detects keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, it generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[0522] Call blocking and notifications

[0523] The server plays the automated message sent by the AI ​​system to the caller and disconnects the call. At the same time, the server adds the phone number to a blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[0524] Continuous learning and updates

[0525] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0526] Prompt Sentence Examples

[0527] "Analyze the content of this call, and if suspicious keywords (such as 'identity verification,' 'account,' 'funds,' etc.) are detected, determine that it may be a scam and generate an automated response message."

[0528] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0529] Step 1:

[0530] The server monitors all incoming calls. When a new call is received, it captures the call information (caller's phone number, date and time, etc.). (Input): New incoming call. (Data processing): Capture and record call information. (Output): Caller information.

[0531] Step 2:

[0532] The server queries the caller's phone number in a database system. (Input): Caller information for incoming calls. (Data calculation): Query the database. (Output): Query results (blacklist match).

[0533] Step 3:

[0534] The server judges the query result and checks whether the caller is on the blacklist. If the caller is on the blacklist, the server will automatically block the call. (Input): Database query result. (Data calculation): Judgment process. (Output): Call blocking.

[0535] Step 4:

[0536] If the caller is not on the blacklist, the server records the call and sends the data to the generation AI system. (Input): Call information that does not fall under the blacklist. (Data processing): Recording of call content and data transmission. (Output): Recorded data.

[0537] Step 5:

[0538] The generative AI system analyzes the received call data and assesses the risk of fraud. (Input): Recorded data. (Data calculation): Analysis of call content and risk assessment using natural language processing. (Output): Risk assessment results (presence or absence of fraud risk and its degree).

[0539] Step 6:

[0540] If the risk of fraud is determined to be high, the AI ​​system generates an automatic response message and sends it to the server. (Input): Risk assessment result. (Data calculation): Generation of automatic response message. (Output): Automatic response message.

[0541] Step 7:

[0542] The server plays the automated response message sent by the AI ​​generation system to the caller and disconnects the call. At the same time, it adds the target phone number to the blacklist. (Input): Auto-response message, caller information. (Data processing): Plays the message and disconnects the call, updates the blacklist. (Output): Phone number added to the blacklist.

[0543] Step 8:

[0544] The user notification system sends real-time notifications to users and their families when fraudulent calls are detected. (Input): Risk assessment results and blacklist update information. (Data calculation): Notification message generation. (Output): Notification message (SMS, notification app, email).

[0545] Step 9:

[0546] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the model of the generative AI system. (Input): Newly detected fraud data. (Data calculation): Update training data and retrain the model. (Output): Improved AI model.

[0547] (Application example 1)

[0548] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0549] With the elderly currently at a higher risk of falling victim to special frauds, there is a need for a system to protect them from telephone fraud. However, existing systems have difficulty analyzing call content in real time and immediately assessing the risk of fraud. Even if calls with a high risk of fraud are detected, appropriate measures may not be taken promptly, making them insufficient to ensure the safety of the elderly. Furthermore, there is a risk that responses to new fraud methods will be delayed, reducing the accuracy of the system. It is necessary to solve these issues and provide an environment in which the elderly can live with peace of mind.

[0550] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0551] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying users and their families in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means operated via a mobile communication terminal with a call monitoring function; and means with a security function for sending the generated automated response message to a designated recipient. This safely protects elderly people from special frauds, analyzes suspicious calls in real time to take appropriate measures, and enables elderly people and their families to quickly understand the risks.

[0552] A "mobile communication terminal with call monitoring function" refers to a portable communication device that has the function of receiving calls and monitoring their content.

[0553] The "means for querying a database for caller information" refers to a system that has the function of querying and confirming caller information from a database that stores information about the caller.

[0554] "Automatic call blocking" is a feature that automatically blocks calls when a suspicious phone number or call is detected.

[0555] "Means for assessing the risk of fraud" refers to a function that analyzes the content of a call and determines the likelihood that the call is fraudulent.

[0556] "Means for generating and sending automatic response messages to callers" refers to a function that automatically creates a pre-defined message and sends it to callers when it is determined that there is a high risk of fraud.

[0557] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function that adds telephone numbers that are determined to be at high risk of fraud to a specific list and manages that list in a database.

[0558] "Means of notifying users and their families in real time" is a function that quickly notifies users and their families of any calls that pose a risk of fraud when such calls are detected.

[0559] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function that allows the system to learn newly discovered fraud methods and patterns, update the analysis model based on them, and improve accuracy.

[0560] The present invention relates to a system for protecting elderly people from special frauds, which is implemented using a mobile communication terminal and a server. Specific embodiments of the system are described below.

[0561] 1. System Configuration

[0562] The system consists of the following main components:

[0563] 1. Mobile communication terminals

[0564] 2. Call Reception Server

[0565] 3. Generative AI Systems

[0566] 4. Database Systems

[0567] 5. User Notification System

[0568] 6. Continuous Learning System

[0569] 2. System Operation

[0570] Mobile communication terminal

[0571] A mobile communication terminal is a portable communication device equipped with a call monitoring function. When a call is received, the terminal records the call and transmits the caller information to a server.

[0572] Call acceptance server

[0573] The server receives calls and checks the caller's phone number against a database. If a call comes from an unknown number, it records the call and sends it to a generative AI system. If the number is suspicious, the call is automatically blocked.

[0574] Generative AI System

[0575] The generative AI system analyzes the content of calls in real time and assesses the risk of fraud. The AI ​​model converts the voice data into text and analyzes the text to determine the risk of fraud. If it determines that the risk of fraud is high, it generates an appropriate automated response message and sends it to the server.

[0576] Database System

[0577] The database system stores and manages information on suspicious phone numbers and fraudulent methods. For example, a blacklist is managed in this database, and the server adds new blacklist items based on information from the generative AI system.

[0578] User Notification System

[0579] The user notification system will send real-time notifications to users and their families when a fraudulent call is detected. This notification will be sent via SMS, a notification app, or email, allowing users and their families to quickly understand the risk.

[0580] Continuous Learning System

[0581] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0582] 3. Hardware and Software

[0583] The following hardware and software are used to implement this system:

[0584] Hardware: Smartphones, servers

[0585] Software: Python, Requests library, smtplib, custom API (call acceptance, generative AI, database management)

[0586] 4. Examples and prompts

[0587] Specific examples

[0588] An elderly person installs a smartphone app. One day, the elderly person receives a call from an unknown number. The app monitors the call and determines it to be suspicious. A notification is sent to the family saying, "There was a suspicious call from 0123456789."

[0589] Prompt example

[0590] "Please explain the functionality of an AI application that analyzes calls from unknown numbers in real time to protect seniors from specialized fraud."

[0591] This will ensure that elderly people are safely protected from special fraud, suspicious calls will be analyzed in real time and appropriate measures will be taken, and elderly people and their families will be able to quickly understand the risks.

[0592] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0593] Step 1:

[0594] The server receives calls from mobile communication terminals, obtains the caller's phone number, and checks the caller information against a database to see if the phone number is suspicious.

[0595] Specifically, it receives the caller's phone number as input, searches for the corresponding phone number in the database, and returns a flag indicating whether the caller's phone number is suspicious.

[0596] Step 2:

[0597] If the server determines that the caller is a suspicious number, it will automatically block the call.

[0598] Specifically, it receives a suspicious phone number flag as input, blocks the call, and returns a confirmation message that the call has been blocked as output.

[0599] Step 3:

[0600] If the caller is not a suspicious phone number, the server records the call and sends the data to the generation AI system.

[0601] Specifically, it receives voice call data as input, sends it to the AI ​​analysis system, and returns a message confirming the transmission of the voice call data as output.

[0602] Step 4:

[0603] The generative AI system analyzes incoming call content in real time and assesses the risk of fraud.

[0604] Specifically, the system receives voice call data as input, performs text analysis using an AI model, and returns the fraud risk assessment results to the server as output.

[0605] Step 5:

[0606] The server receives the evaluation results from the generative AI system and, if it determines that there is a high risk of fraud, generates an automated response message to send to the caller.

[0607] Specifically, it receives the fraud risk assessment result as input, generates an automatic response message, and returns the response message to be sent to the caller as output.

[0608] Step 6:

[0609] The server adds phone numbers that are deemed to pose a fraud risk to a blacklist in its database, automatically blocking future calls.

[0610] Specifically, it receives a fraudulent phone number as input, registers it in a blacklist in the database, and returns a message confirming the registration to the blacklist as output.

[0611] Step 7:

[0612] The user notification system notifies users and their families in real time when calls that pose a risk of fraud are detected.

[0613] It takes the fraud risk assessment result as input, generates notification data, and sends notifications via SMS, notification app, or email as output.

[0614] Step 8:

[0615] The continuous learning system stores newly detected fraud patterns as training data and periodically updates the generative AI system's model.

[0616] Specifically, it receives new fraud pattern data as input, stores it in the training database, and returns a confirmation message for the updated AI model as output.

[0617] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0618] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds, and in particular, improves the accuracy of fraud detection by combining it with an emotion engine that recognizes the user's emotions. The following describes specific embodiments of the present invention.

[0619] System Overview

[0620] The system consists of the following main components:

[0621] 1. Call Reception Server

[0622] 2. Generative AI Systems

[0623] 3. Database System

[0624] 4. User Notification System

[0625] 5. Continuous Learning System

[0626] 6. Emotion Engine

[0627] Program processing

[0628] Call acceptance server

[0629] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[0630] Generative AI System

[0631] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0632] Database System

[0633] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[0634] User Notification System

[0635] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0636] Continuous Learning System

[0637] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0638] Emotion Engine

[0639] The emotion engine analyzes the user's emotions during the call and sends the results to the generative AI system. The emotion engine analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc.

[0640] Specific examples

[0641] Receiving and analyzing calls

[0642] For example, if a call comes in to a senior citizen's phone from an unknown number, the server will record the call and query the caller's phone number against a database. If the database does not match any blacklist, the call will be sent to the generative AI system and emotion engine.

[0643] Collaboration between analysis and generative AI using an emotion engine

[0644] The generative AI system analyzes the call content in real time, detecting keywords such as "identity verification," "account," and "funds." At the same time, the emotion engine analyzes the voice and evaluates the user's emotional state. If the emotion engine determines that the user's emotions are abnormal, such as anxiety or fear, the generative AI will take this into account when assessing the fraud risk.

[0645] Auto-response and blacklist management

[0646] If the AI ​​system determines that there is a high risk of fraud, it generates an automated response message such as "We are currently unavailable. We will contact you shortly" and sends it to the server. The server plays this message to the caller and disconnects the call. The server then adds the caller's phone number to a blacklist in its database.

[0647] Notification and follow-up

[0648] Based on the analysis results of the emotion engine and the risk assessment of the generative AI, the user notification system will send warning messages such as "A suspicious call has been detected" to the elderly person's family members, and will also provide further detailed advice and follow-up notifications if necessary.

[0649] Continuous learning and updates

[0650] The continuous learning system saves the detected fraud patterns and data from the emotion engine as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving the fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0651] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[0652] The processing flow will be explained below.

[0653] Step 1:

[0654] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[0655] Step 2:

[0656] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[0657] Step 3:

[0658] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[0659] Step 4:

[0660] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generative AI system and emotion engine.

[0661] Step 5:

[0662] The generative AI system analyzes the transmitted call data and assesses the risk of specialized fraud. The analysis involves detecting keywords and conversation patterns, such as "identity verification," "account," and "funds."

[0663] Step 6:

[0664] The emotion engine analyzes voice data during a call to detect the user's tone and emotional fluctuations, thereby assessing whether the user is experiencing emotions such as anxiety, fear, or anger.

[0665] Step 7:

[0666] The generative AI system adjusts the fraud risk assessment based on the sentiment analysis results from the emotion engine. For example, if the emotion engine detects anxiety or fear in the user, it will determine that the fraud risk is high.

[0667] Step 8:

[0668] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0669] Step 9:

[0670] The server plays the automated response message sent by the AI ​​system to the caller, disconnects the call, and adds the caller's phone number to a blacklist in the database.

[0671] Step 10:

[0672] The user notification system will send real-time notifications to users and their families based on the analysis results of the emotion engine and the risk assessment of the generative AI system via SMS, notification app, or email.

[0673] Step 11:

[0674] The notification message will include something like, "A suspicious call has been detected. Please contact your system administrator for more information." In some cases, more detailed advice or follow-up notifications will be provided.

[0675] Step 12:

[0676] The continuous learning system saves detected fraud patterns and emotion engine data as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving fraud detection accuracy in future iterations.

[0677] Example 2

[0678] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0679] There is a problem that exposes elderly people to the risk of special fraud. Existing telephone systems may not be able to detect fraud risks, increasing the likelihood that elderly people will become victims of fraud. In addition, it is difficult to quickly notify and respond to fraud risks, so prompt measures are needed to minimize damage.

[0680] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for analyzing the user's emotions and reflecting them in the fraud risk assessment; means for generating and sending an automatic response message to the caller if the fraud risk is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call with a risk of fraud is detected; and means for learning newly detected fraud patterns and improving the accuracy of the analysis model. This protects elderly people from fraud, enables rapid and accurate detection of fraud risks, and enables appropriate responses.

[0681] "Means for receiving calls and checking the database for caller information" is a function that detects an incoming call and checks the caller's telephone number against a database to confirm it.

[0682] "Means to automatically block calls if the caller is a suspicious phone number" is a function that automatically prevents calls from being received if the caller's phone number is included in a blacklist in the database.

[0683] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of a conversation during a call in real time and determines the risk of fraud based on that content.

[0684] "Means for analyzing user emotions and reflecting them in fraud risk assessment" is a function that analyzes changes in the user's tone of voice and emotions during a call and uses the results to help assess fraud risk.

[0685] "Means for generating and sending an automatic response message to a caller when a high risk of fraud is determined" is a function that sends an automatically created message to a caller for calls that are assessed as having a high risk of fraud.

[0686] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for registering telephone numbers that are determined to pose a risk of fraud in a database and for retaining and managing that information.

[0687] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that notifies users and their families in real time when a call that poses a high risk of fraud occurs.

[0688] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that learns newly discovered fraud methods and patterns and improves the accuracy of the fraud detection model for the entire system.

[0689] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds. In particular, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of fraud detection is improved. This system is realized using hardware and software. The following describes in detail an embodiment of the present invention.

[0690] System Overview

[0691] The system consists of the following main components:

[0692] 1. Call Reception Server

[0693] 2. Generative AI Systems

[0694] 3. Database System

[0695] 4. User Notification System

[0696] 5. Continuous Learning System

[0697] 6. Emotion Engine

[0698] Call acceptance server

[0699] The server monitors all incoming calls, and if a call comes from an unknown phone number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the generation AI system. The hardware used is a standard server and telephone system for call management. The software used is call recording software and database query software.

[0700] Examples:

[0701] When an elderly user receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to a generative AI system and emotion engine.

[0702] Generative AI System

[0703] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server. A general language model is used as the generative AI model.

[0704] Examples:

[0705] The generative AI system analyzes call content in real time, detecting keywords such as "identity verification," "account," and "funds," while also taking into account the analysis results of the emotion engine to assess fraud risk.

[0706] Example prompt sentence:

[0707] "Assess your risk of fraud based on the following conversation: 'Hello, this is from your bank and I need to verify your account.'"

[0708] Database System

[0709] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the generative AI system deems suspicious to a blacklist in the database, automatically blocking future calls. The hardware used is a standard database server, and the software is a common database management system.

[0710] Examples:

[0711] The server adds the phone number of callers it deems a high risk of fraud to a database and automatically blocks future calls.

[0712] User Notification System

[0713] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. The hardware used is a communication server for sending notifications, and the software used is a notification app or SMS sending system.

[0714] Examples:

[0715] Based on the analysis results of the emotion engine and the risk assessment by the generating AI, a warning message such as "A suspicious call has been detected" is sent via SMS to the elderly person's family.

[0716] Continuous Learning System

[0717] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0718] Examples:

[0719] Newly detected fraud patterns, along with data from the emotion engine, are saved as training data to update the model of the generative AI system.

[0720] Emotion Engine

[0721] The emotion engine analyzes the user's emotions during a call and sends the results to the generative AI system. It analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc. The hardware uses a computer for voice analysis, and the software uses an emotion analysis algorithm.

[0722] Examples:

[0723] The call content is analyzed, and if the user's emotions indicate anxiety or fear, the results are sent to a generative AI system, which takes this into account when assessing fraud risk.

[0724] Example prompt sentence:

[0725] "Analyze the emotions from this audio data and tell us what emotions the user is feeling."

[0726] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[0727] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0728] Step 1:

[0729] The server detects incoming phone calls and monitors them. If an incoming call comes from an unknown phone number, the server records the call and records details such as the call timestamp, the caller's phone number, and the call length. Specifically, when an elderly user receives a call from an unknown number on their device, the server begins recording the call. The input is the call details, and the output is the recorded audio data.

[0730] Step 2:

[0731] The server queries the caller information of the recorded call in a database system to see if the number is on the blacklist. Specifically, the server queries the caller's phone number in the database and gets the result "not on the blacklist." The input is the caller's phone number, and the output is the database query result.

[0732] Step 3:

[0733] If the server confirms that the caller is not on the blacklist, it records the call and sends the data to the generative AI system and emotion engine. Specifically, after the call is recorded, the server sends the audio data to the generative AI system and emotion engine. The input is the audio data of the recorded call, and the output is the transfer of data to the generative AI system and emotion engine.

[0734] Step 4:

[0735] The generative AI system analyzes the received voice data and detects specific keywords and patterns. The emotion engine also analyzes the voice data and evaluates the user's emotional state. Specifically, the generative AI system detects keywords such as "identity verification," "account," and "funds" from the call, and the emotion engine analyzes the tone and speed of the user's voice to determine whether they are feeling anxious. The input is the voice data of the call content, and the output is specific keywords and the user's emotional evaluation result.

[0736] Step 5:

[0737] The server evaluates whether the risk of fraud is high based on the fraud risk assessment results provided by the generative AI system and the analysis results of the emotion engine. Specifically, if the generative AI system returns a high risk score and the emotion engine concludes that the user is anxious, the server determines that the risk of fraud is high. The inputs are the assessment results of the generative AI system and the analysis results of the emotion engine, and the output is the risk assessment result.

[0738] Step 6:

[0739] If the server determines that there is a high risk of fraud, it sends the generated auto-answer message to the caller and disconnects the call.Specifically, a message such as "There is no one in charge at the moment. We will contact you later" is played and the call is disconnected.The input is a template for the auto-answer message, and the output is sending a message to the caller and disconnecting the call.

[0740] Step 7:

[0741] The server adds the caller's phone number to a blacklist in the database system. Specifically, the server registers the caller's phone number in the database, and future calls are automatically blocked. The input is the caller's phone number, and the output is the registration result in the database.

[0742] Step 8:

[0743] The user notification system notifies users and their families in real time about detected fraud risks. Specifically, it sends a warning message to the elderly family member's smartphone via SMS saying, "A suspicious call has been detected." The input is the fraud risk assessment result, and the output is the sending of a notification message.

[0744] Step 9:

[0745] The continuous learning system saves newly detected fraud patterns and emotion data as learning data and periodically updates the generative AI system and emotion engine models. Specifically, the detected fraud patterns and emotion data are saved in a database, and the model is periodically updated. The input is the newly detected fraud patterns and emotion data, and the output is the updated AI model.

[0746] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[0747] (Application example 2)

[0748] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0749] In recent years, the number of elderly people who fall victim to special frauds over the phone has been increasing. However, conventional fraud prevention systems are finding it difficult to respond effectively to the increasing diversity and sophistication of fraud methods. In addition, since it is difficult for elderly people to judge the content of phone calls themselves, there is a need for a system that can prevent fraud damage before it occurs. In particular, the development of an advanced fraud prevention system that automatically responds to fraud methods and combines real-time emotion analysis is a challenge.

[0750] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automatic response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; emotion analysis means for evaluating the user's emotional state; and means for complementing the fraud risk assessment based on the user's emotional state. This significantly reduces the risk of elderly people falling victim to special telephone frauds and provides an environment in which they can live safely.

[0751] "Means for receiving calls and checking the caller information against a database" is a function that automatically obtains information about an incoming call and compares the caller information with a pre-registered database.

[0752] "Means for automatically blocking calls when the caller is a suspicious phone number" is a function that automatically blocks incoming calls from suspicious phone numbers based on the results of a database match.

[0753] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of the conversation in real time during a call and assesses the possibility of fraud based on that content.

[0754] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" is a function that automatically generates a response message and sends it to the caller when the system determines that there is a high risk of fraud.

[0755] "Means for storing and managing blacklisted telephone numbers in a database" refers to a management function that stores telephone numbers of callers deemed suspicious in a database and consistently blocks subsequent calls.

[0756] "Means of notifying users and their families in real time when calls that pose a high risk of fraud are detected" is a function that immediately sends a warning to users and their families when calls that pose a high risk of fraud are detected.

[0757] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that adds newly discovered fraud methods and patterns as learning data and improves the accuracy of the analysis model.

[0758] The "emotion analysis means for evaluating the user's emotional state" is a function that analyzes the user's tone of voice and choice of words during a call and evaluates their emotional state.

[0759] "Means to complement fraud risk assessment based on emotional state" is a function that further refines fraud risk assessment based on the results of user emotional analysis.

[0760] To implement the present invention, a system including the following procedures and components is used.

[0761] System Overview

[0762] The system consists of a server that receives and analyzes calls in real time, a terminal that monitors the user's call status, and a notification system that notifies users and their families based on the results of the call analysis.

[0763] 1. Receiving calls and querying caller information

[0764] The server automatically records calls received by the user and checks the caller information against a database that includes a blacklist of suspicious phone numbers.

[0765] 2. Real-time analysis and automatic response

[0766] The server analyzes the call content in real time and assesses the risk of fraud based on the content of the conversation and the keywords used.

[0767] If the risk of fraud is determined to be high, the server generates an automated response message and sends it to the caller, designed to prevent fraud before it occurs.

[0768] 3. Emotion analysis

[0769] The server uses an emotion engine to analyze the tone, rate and emotional fluctuations of the user's voice during the call to assess the user's emotional state.

[0770] Sentiment analysis results are used to complement fraud risk assessment.

[0771] 4. Notification System

[0772] If a fraudulent call is detected, the notification system will send a real-time notification to the user and their family via SMS, notification app, or email.

[0773] 5. Blacklist Management

[0774] The server stores phone numbers that are deemed suspicious in a blacklist in its database and automatically blocks future calls.

[0775] 6. Continuous Learning System

[0776] The continuous learning system saves newly detected fraud patterns as training data to improve the accuracy of the analysis model.

[0777] Hardware and software used

[0778] 1. Hardware

[0779] Server: Data processing device for analysis and notification

[0780] Smartphone: Records user calls and sends them to a server

[0781] 2. Software

[0782] Call reception app: Records user calls and sends the data to a server

[0783] Generative AI system: Analyzes call content and assesses fraud risk

[0784] Emotion Engine: Analyzes the user's emotional state

[0785] Notification app: Sends notifications about fraud risks

[0786] Specific examples

[0787] 1. Receiving and analyzing calls

[0788] When a user receives a call from an unknown number, the server records the call and checks the caller's number against a database. If the call is not blacklisted, the server analyzes the call in real time.

[0789] 2. Sentiment Analysis and Risk Assessment

[0790] The server uses an emotion engine to analyze the user's emotional state, and the generative AI system analyzes keywords in the call content. The results of the emotion analysis and the call content analysis are combined to assess the fraud risk.

[0791] 3. Auto-replies and notifications

[0792] If the risk of fraud is deemed high, the server generates an automated response message and sends it to the caller, while simultaneously sending real-time notifications to the user and their family members via SMS, notification app, or email.

[0793] Prompt Sentence Examples

[0794] Call content: {Call text}

[0795] Emotional state: {User emotion}

[0796] Criteria: Assess the fraud risk and return a "High Fraud Risk" message if the risk is high.

[0797] In this way, this system reduces the risk of users becoming victims of special fraud and provides an environment in which users can make calls with peace of mind.

[0798] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0799] Step 1:

[0800] When a user receives a call, the device records the call and obtains the caller's phone number. The obtained call data and caller information are sent to the server. The input is the incoming call's voice data and the caller's phone number, and the output is that the data is sent to the server.

[0801] Step 2:

[0802] The server checks the received caller information against the database. If the check finds that the caller is registered on the blacklist, the server automatically blocks the call. The input is the caller's phone number, and the output is the check result and the call blocking process.

[0803] Step 3:

[0804] If the caller is not on the blacklist, the server analyzes the call in real time using a generative AI system that performs keyword analysis and pattern matching on the call content. The input is the audio data of the call, and the output is the fraud risk assessment result.

[0805] Step 4:

[0806] The server uses an emotion engine to analyze the tone, speed, and emotional fluctuations of the user's voice during the call to evaluate the user's emotional state. The input is the voice data of the call, and the output is the evaluation result of the user's emotional state.

[0807] Step 5:

[0808] The server evaluates the risk of fraud by combining the results of keyword analysis by the generative AI system and the results of the emotional state evaluation by the emotion engine. If the risk of fraud is determined to be high, the server generates an automatic response message and sends it to the caller. The input is the analysis results of the generative AI system and the evaluation results of the emotion engine, and the output is the automatic response message.

[0809] Step 6:

[0810] If a fraud risk is detected, the server sends a notification to the user and their family members via SMS, a notification app, or email. The input is the fraud risk assessment result, and the output is the notification message.

[0811] Step 7:

[0812] The server adds any phone numbers it deems suspicious to a blacklist in its database, automatically blocking future calls. The input is the suspicious phone number, and the output is the registration status in the database.

[0813] Step 8:

[0814] The continuous learning system improves accuracy by saving newly detected fraud patterns as training data and updating the analytical models of the generative AI system and emotion engine. The input is the newly detected fraud patterns, and the output is the updated models.

[0815] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0816] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0817] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0818] [Third embodiment]

[0819] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0820] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0821] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0822] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0823] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0824] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0825] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0826] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0827] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0828] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0829] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0830] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0831] The present invention is an AI telephone handling service system for helping elderly people avoid falling victim to special fraud, and can be implemented as follows.

[0832] System Overview

[0833] The system consists of the following main components:

[0834] 1. Call Reception Server

[0835] 2. Generative AI Systems

[0836] 3. Database System

[0837] 4. User Notification System

[0838] 5. Continuous Learning System

[0839] Program processing

[0840] Call acceptance server

[0841] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[0842] Generative AI System

[0843] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0844] Database System

[0845] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[0846] User Notification System

[0847] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0848] Continuous Learning System

[0849] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0850] Specific examples

[0851] Receiving and analyzing calls

[0852] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the database for the caller's phone number. If the call is not blacklisted, the server sends the call to the AI ​​system.

[0853] Analysis and automatic response by generative AI

[0854] The AI ​​system analyzes the content of calls in real time, detecting keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, the AI ​​generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[0855] Call blocking and notifications

[0856] The server plays the automated response message sent by the AI ​​system to the caller and disconnects the call. At the same time, the target phone number is added to the blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[0857] Continuous learning and updates

[0858] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0859] In this way, this system can protect the elderly from the risk of special fraud and provide an environment where they can live in peace of mind.

[0860] The processing flow will be explained below.

[0861] Step 1:

[0862] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[0863] Step 2:

[0864] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[0865] Step 3:

[0866] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[0867] Step 4:

[0868] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generation AI system.

[0869] Step 5:

[0870] The generative AI system analyzes incoming call data to assess fraud risk, detecting keywords and conversation patterns such as "identity verification," "account," and "funds."

[0871] Step 6:

[0872] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[0873] Step 7:

[0874] The server plays the automated response message sent by the generating AI system to the caller and then disconnects the call.

[0875] Step 8:

[0876] The server adds the caller's phone number to a blacklist in its database, automatically blocking future calls.

[0877] Step 9:

[0878] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected via SMS, notification app, or email.

[0879] Step 10:

[0880] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the generative AI system's model, thereby improving the analysis accuracy of the entire system.

[0881] Example 1

[0882] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0883] In modern society, the risk of elderly people becoming victims of special frauds is increasing. Such frauds are becoming more sophisticated, with a particularly rapid increase in fraud methods over the phone. Elderly people have difficulty recognizing fraud and are often more susceptible to falling victim. Therefore, there is a need for a system that can automatically assess fraud risks and respond quickly. Furthermore, there is a need for a means to notify elderly people and their families of fraud risks in real time, and a system that can continuously respond to new fraud patterns is also needed.

[0884] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0885] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means for recording the content of calls and sending it to a generative AI model system for analysis; and means for generating an automated response message based on the analysis results and sending it to the server. This protects elderly people from the risk of special fraud and allows them to receive prompt fraud warnings. It also enables continuous learning and improvement of the accuracy of the analysis model to respond to new fraud techniques.

[0886] "Means for receiving calls and checking the database for caller information" refers to a function that accepts external telephone calls and checks the caller's telephone number against an internal database to see if the caller matches the registered information.

[0887] "Means for automatically blocking calls when the caller is a suspicious phone number" refers to a function that automatically disconnects the call and prevents the call from being forwarded if the caller's phone number is blacklisted as a result of a database check.

[0888] "A means of analyzing the content of calls in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that uses voice recognition to instantly analyze the content of calls and determine whether there is a possibility of fraud if the call is not on the blacklist.

[0889] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" refers to a function that automatically generates a standard response message and sends it to the caller when the analysis of the content of the call indicates a high risk of fraud.

[0890] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for recording telephone numbers of suspicious callers in a database and continuously managing and updating that information.

[0891] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that sends an immediate warning notification to users and their families when a call that poses a high risk of fraud is detected.

[0892] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function for improving the accuracy of fraud detection by accumulating newly detected fraud methods and patterns as learning data and periodically updating the analysis algorithm model.

[0893] "Means for recording the contents of a call and transmitting it to a generative AI model system for analysis" refers to a function that records the audio during a call and transmits the recorded data to an analysis system that uses a generative AI model, thereby performing processing to analyze the contents of the call.

[0894] "Means for generating an automatic response message based on the analysis results and sending it to the server" refers to a function that creates an appropriate automatic response message based on the results of analysis by the generative AI model system and sends that message to the server.

[0895] MODE FOR CARRYING OUT THE INVENTION

[0896] This invention is an AI telephone handling service system designed to protect elderly people from falling victim to special fraud. This system consists of the following main components: a call reception server, a generation AI system, a database system, a user notification system, and a continuous learning system. Below, we provide a detailed explanation of each component and explain how the system works.

[0897] System Overview

[0898] Call acceptance server

[0899] The call reception server monitors all incoming calls. When a new call comes in, it captures the call information and checks whether it is an unknown phone number. The server then queries the caller's phone number in a database system to determine whether the number is blacklisted. If the caller is not blacklisted, it records the call and sends the data to the generation AI system.

[0900] Examples of hardware and software used: VoIP devices, communication control servers

[0901] Generative AI System

[0902] The generative AI system analyzes incoming call data and assesses the risk of fraud. The system includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an automated response message and sends it to the server.

[0903] Examples of hardware and software used: Natural Language Processing (NLP) engine, Python, TensorFlow library

[0904] Database System

[0905] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the AI ​​system determines to be suspicious to a blacklist in the database, automatically blocking future calls.

[0906] Examples of hardware and software used: SQL database server

[0907] User Notification System

[0908] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[0909] Examples of hardware and software used: notification server, SMS gateway, notification application

[0910] Continuous Learning System

[0911] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0912] Examples of hardware and software used: machine learning platform, data storage system

[0913] Specific examples

[0914] Receiving and analyzing calls

[0915] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to the AI ​​system.

[0916] Analysis and automatic response by generative AI

[0917] The AI ​​generation system analyzes the content of calls in real time. For example, it detects keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, it generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[0918] Call blocking and notifications

[0919] The server plays the automated message sent by the AI ​​system to the caller and disconnects the call. At the same time, the server adds the phone number to a blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[0920] Continuous learning and updates

[0921] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[0922] Prompt Sentence Examples

[0923] "Analyze the content of this call, and if suspicious keywords (such as 'identity verification,' 'account,' 'funds,' etc.) are detected, determine that it may be a scam and generate an automated response message."

[0924] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0925] Step 1:

[0926] The server monitors all incoming calls. When a new call is received, it captures the call information (caller's phone number, date and time, etc.). (Input): New incoming call. (Data processing): Capture and record call information. (Output): Caller information.

[0927] Step 2:

[0928] The server queries the caller's phone number in a database system. (Input): Caller information for incoming calls. (Data calculation): Query the database. (Output): Query results (blacklist match).

[0929] Step 3:

[0930] The server judges the query result and checks whether the caller is on the blacklist. If the caller is on the blacklist, the server will automatically block the call. (Input): Database query result. (Data calculation): Judgment process. (Output): Call blocking.

[0931] Step 4:

[0932] If the caller is not on the blacklist, the server records the call and sends the data to the generation AI system. (Input): Call information that does not fall under the blacklist. (Data processing): Recording of call content and data transmission. (Output): Recorded data.

[0933] Step 5:

[0934] The generative AI system analyzes the received call data and assesses the risk of fraud. (Input): Recorded data. (Data calculation): Analysis of call content and risk assessment using natural language processing. (Output): Risk assessment results (presence or absence of fraud risk and its degree).

[0935] Step 6:

[0936] If the risk of fraud is determined to be high, the AI ​​system generates an automatic response message and sends it to the server. (Input): Risk assessment result. (Data calculation): Generation of automatic response message. (Output): Automatic response message.

[0937] Step 7:

[0938] The server plays the automated response message sent by the AI ​​generation system to the caller and disconnects the call. At the same time, it adds the target phone number to the blacklist. (Input): Auto-response message, caller information. (Data processing): Plays the message and disconnects the call, updates the blacklist. (Output): Phone number added to the blacklist.

[0939] Step 8:

[0940] The user notification system sends real-time notifications to users and their families when fraudulent calls are detected. (Input): Risk assessment results and blacklist update information. (Data calculation): Notification message generation. (Output): Notification message (SMS, notification app, email).

[0941] Step 9:

[0942] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the model of the generative AI system. (Input): Newly detected fraud data. (Data calculation): Update training data and retrain the model. (Output): Improved AI model.

[0943] (Application example 1)

[0944] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0945] With the elderly currently at a higher risk of falling victim to special frauds, there is a need for a system to protect them from telephone fraud. However, existing systems have difficulty analyzing call content in real time and immediately assessing the risk of fraud. Even if calls with a high risk of fraud are detected, appropriate measures may not be taken promptly, making them insufficient to ensure the safety of the elderly. Furthermore, there is a risk that responses to new fraud methods will be delayed, reducing the accuracy of the system. It is necessary to solve these issues and provide an environment in which the elderly can live with peace of mind.

[0946] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0947] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying users and their families in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means operated via a mobile communication terminal with a call monitoring function; and means with a security function for sending the generated automated response message to a designated recipient. This safely protects elderly people from special frauds, analyzes suspicious calls in real time to take appropriate measures, and enables elderly people and their families to quickly understand the risks.

[0948] A "mobile communication terminal with call monitoring function" refers to a portable communication device that has the function of receiving calls and monitoring their content.

[0949] The "means for querying a database for caller information" refers to a system that has the function of querying and confirming caller information from a database that stores information about the caller.

[0950] "Automatic call blocking" is a feature that automatically blocks calls when a suspicious phone number or call is detected.

[0951] "Means for assessing the risk of fraud" refers to a function that analyzes the content of a call and determines the likelihood that the call is fraudulent.

[0952] "Means for generating and sending automatic response messages to callers" refers to a function that automatically creates a pre-defined message and sends it to callers when it is determined that there is a high risk of fraud.

[0953] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function that adds telephone numbers that are determined to be at high risk of fraud to a specific list and manages that list in a database.

[0954] "Means of notifying users and their families in real time" is a function that quickly notifies users and their families of any calls that pose a risk of fraud when such calls are detected.

[0955] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function that allows the system to learn newly discovered fraud methods and patterns, update the analysis model based on them, and improve accuracy.

[0956] The present invention relates to a system for protecting elderly people from special frauds, which is implemented using a mobile communication terminal and a server. Specific embodiments of the system are described below.

[0957] 1. System Configuration

[0958] The system consists of the following main components:

[0959] 1. Mobile communication terminals

[0960] 2. Call Reception Server

[0961] 3. Generative AI Systems

[0962] 4. Database Systems

[0963] 5. User Notification System

[0964] 6. Continuous Learning System

[0965] 2. System Operation

[0966] Mobile communication terminal

[0967] A mobile communication terminal is a portable communication device equipped with a call monitoring function. When a call is received, the terminal records the call and transmits the caller information to a server.

[0968] Call acceptance server

[0969] The server receives calls and checks the caller's phone number against a database. If a call comes from an unknown number, it records the call and sends it to a generative AI system. If the number is suspicious, the call is automatically blocked.

[0970] Generative AI System

[0971] The generative AI system analyzes the content of calls in real time and assesses the risk of fraud. The AI ​​model converts the voice data into text and analyzes the text to determine the risk of fraud. If it determines that the risk of fraud is high, it generates an appropriate automated response message and sends it to the server.

[0972] Database System

[0973] The database system stores and manages information on suspicious phone numbers and fraudulent methods. For example, a blacklist is managed in this database, and the server adds new blacklist items based on information from the generative AI system.

[0974] User Notification System

[0975] The user notification system will send real-time notifications to users and their families when a fraudulent call is detected. This notification will be sent via SMS, a notification app, or email, allowing users and their families to quickly understand the risk.

[0976] Continuous Learning System

[0977] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[0978] 3. Hardware and Software

[0979] The following hardware and software are used to implement this system:

[0980] Hardware: Smartphones, servers

[0981] Software: Python, Requests library, smtplib, custom API (call acceptance, generative AI, database management)

[0982] 4. Examples and prompts

[0983] Specific examples

[0984] An elderly person installs a smartphone app. One day, the elderly person receives a call from an unknown number. The app monitors the call and determines it to be suspicious. A notification is sent to the family saying, "There was a suspicious call from 0123456789."

[0985] Prompt example

[0986] "Please explain the functionality of an AI application that analyzes calls from unknown numbers in real time to protect seniors from specialized fraud."

[0987] This will ensure that elderly people are safely protected from special fraud, suspicious calls will be analyzed in real time and appropriate measures will be taken, and elderly people and their families will be able to quickly understand the risks.

[0988] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0989] Step 1:

[0990] The server receives calls from mobile communication terminals, obtains the caller's phone number, and checks the caller information against a database to see if the phone number is suspicious.

[0991] Specifically, it receives the caller's phone number as input, searches for the corresponding phone number in the database, and returns a flag indicating whether the caller's phone number is suspicious.

[0992] Step 2:

[0993] If the server determines that the caller is a suspicious number, it will automatically block the call.

[0994] Specifically, it receives a suspicious phone number flag as input, blocks the call, and returns a confirmation message that the call has been blocked as output.

[0995] Step 3:

[0996] If the caller is not a suspicious phone number, the server records the call and sends the data to the generation AI system.

[0997] Specifically, it receives voice call data as input, sends it to the AI ​​analysis system, and returns a message confirming the transmission of the voice call data as output.

[0998] Step 4:

[0999] The generative AI system analyzes incoming call content in real time and assesses the risk of fraud.

[1000] Specifically, the system receives voice call data as input, performs text analysis using an AI model, and returns the fraud risk assessment results to the server as output.

[1001] Step 5:

[1002] The server receives the evaluation results from the generative AI system and, if it determines that there is a high risk of fraud, generates an automated response message to send to the caller.

[1003] Specifically, it receives the fraud risk assessment result as input, generates an automatic response message, and returns the response message to be sent to the caller as output.

[1004] Step 6:

[1005] The server adds phone numbers that are deemed to pose a fraud risk to a blacklist in its database, automatically blocking future calls.

[1006] Specifically, it receives a fraudulent phone number as input, registers it in a blacklist in the database, and returns a message confirming the registration to the blacklist as output.

[1007] Step 7:

[1008] The user notification system notifies users and their families in real time when calls that pose a risk of fraud are detected.

[1009] It takes the fraud risk assessment result as input, generates notification data, and sends notifications via SMS, notification app, or email as output.

[1010] Step 8:

[1011] The continuous learning system stores newly detected fraud patterns as training data and periodically updates the generative AI system's model.

[1012] Specifically, it receives new fraud pattern data as input, stores it in the training database, and returns a confirmation message for the updated AI model as output.

[1013] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1014] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds, and in particular, improves the accuracy of fraud detection by combining it with an emotion engine that recognizes the user's emotions. The following describes specific embodiments of the present invention.

[1015] System Overview

[1016] The system consists of the following main components:

[1017] 1. Call Reception Server

[1018] 2. Generative AI Systems

[1019] 3. Database System

[1020] 4. User Notification System

[1021] 5. Continuous Learning System

[1022] 6. Emotion Engine

[1023] Program processing

[1024] Call acceptance server

[1025] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[1026] Generative AI System

[1027] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[1028] Database System

[1029] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[1030] User Notification System

[1031] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[1032] Continuous Learning System

[1033] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[1034] Emotion Engine

[1035] The emotion engine analyzes the user's emotions during the call and sends the results to the generative AI system. The emotion engine analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc.

[1036] Specific examples

[1037] Receiving and analyzing calls

[1038] For example, if a call comes in to a senior citizen's phone from an unknown number, the server will record the call and query the caller's phone number against a database. If the database does not match any blacklist, the call will be sent to the generative AI system and emotion engine.

[1039] Collaboration between analysis and generative AI using an emotion engine

[1040] The generative AI system analyzes the call content in real time, detecting keywords such as "identity verification," "account," and "funds." At the same time, the emotion engine analyzes the voice and evaluates the user's emotional state. If the emotion engine determines that the user's emotions are abnormal, such as anxiety or fear, the generative AI will take this into account when assessing the fraud risk.

[1041] Auto-response and blacklist management

[1042] If the AI ​​system determines that there is a high risk of fraud, it generates an automated response message such as "We are currently unavailable. We will contact you shortly" and sends it to the server. The server plays this message to the caller and disconnects the call. The server then adds the caller's phone number to a blacklist in its database.

[1043] Notification and follow-up

[1044] Based on the analysis results of the emotion engine and the risk assessment of the generative AI, the user notification system will send warning messages such as "A suspicious call has been detected" to the elderly person's family members, and will also provide further detailed advice and follow-up notifications if necessary.

[1045] Continuous learning and updates

[1046] The continuous learning system saves the detected fraud patterns and data from the emotion engine as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving the fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[1047] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[1048] The processing flow will be explained below.

[1049] Step 1:

[1050] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[1051] Step 2:

[1052] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[1053] Step 3:

[1054] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[1055] Step 4:

[1056] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generative AI system and emotion engine.

[1057] Step 5:

[1058] The generative AI system analyzes the transmitted call data and assesses the risk of specialized fraud. The analysis involves detecting keywords and conversation patterns, such as "identity verification," "account," and "funds."

[1059] Step 6:

[1060] The emotion engine analyzes voice data during a call to detect the user's tone and emotional fluctuations, thereby assessing whether the user is experiencing emotions such as anxiety, fear, or anger.

[1061] Step 7:

[1062] The generative AI system adjusts the fraud risk assessment based on the sentiment analysis results from the emotion engine. For example, if the emotion engine detects anxiety or fear in the user, it will determine that the fraud risk is high.

[1063] Step 8:

[1064] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[1065] Step 9:

[1066] The server plays the automated response message sent by the AI ​​system to the caller, disconnects the call, and adds the caller's phone number to a blacklist in the database.

[1067] Step 10:

[1068] The user notification system will send real-time notifications to users and their families based on the analysis results of the emotion engine and the risk assessment of the generative AI system via SMS, notification app, or email.

[1069] Step 11:

[1070] The notification message will include something like, "A suspicious call has been detected. Please contact your system administrator for more information." In some cases, more detailed advice or follow-up notifications will be provided.

[1071] Step 12:

[1072] The continuous learning system saves detected fraud patterns and emotion engine data as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving fraud detection accuracy in future iterations.

[1073] Example 2

[1074] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1075] There is a problem that exposes elderly people to the risk of special fraud. Existing telephone systems may not be able to detect fraud risks, increasing the likelihood that elderly people will become victims of fraud. In addition, it is difficult to quickly notify and respond to fraud risks, so prompt measures are needed to minimize damage.

[1076] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for analyzing the user's emotions and reflecting them in the fraud risk assessment; means for generating and sending an automatic response message to the caller if the fraud risk is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call with a risk of fraud is detected; and means for learning newly detected fraud patterns and improving the accuracy of the analysis model. This protects elderly people from fraud, enables rapid and accurate detection of fraud risks, and enables appropriate responses.

[1077] "Means for receiving calls and checking the database for caller information" is a function that detects an incoming call and checks the caller's telephone number against a database to confirm it.

[1078] "Means to automatically block calls if the caller is a suspicious phone number" is a function that automatically prevents calls from being received if the caller's phone number is included in a blacklist in the database.

[1079] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of a conversation during a call in real time and determines the risk of fraud based on that content.

[1080] "Means for analyzing user emotions and reflecting them in fraud risk assessment" is a function that analyzes changes in the user's tone of voice and emotions during a call and uses the results to help assess fraud risk.

[1081] "Means for generating and sending an automatic response message to a caller when a high risk of fraud is determined" is a function that sends an automatically created message to a caller for calls that are assessed as having a high risk of fraud.

[1082] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for registering telephone numbers that are determined to pose a risk of fraud in a database and for retaining and managing that information.

[1083] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that notifies users and their families in real time when a call that poses a high risk of fraud occurs.

[1084] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that learns newly discovered fraud methods and patterns and improves the accuracy of the fraud detection model for the entire system.

[1085] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds. In particular, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of fraud detection is improved. This system is realized using hardware and software. The following describes in detail an embodiment of the present invention.

[1086] System Overview

[1087] The system consists of the following main components:

[1088] 1. Call Reception Server

[1089] 2. Generative AI Systems

[1090] 3. Database System

[1091] 4. User Notification System

[1092] 5. Continuous Learning System

[1093] 6. Emotion Engine

[1094] Call acceptance server

[1095] The server monitors all incoming calls, and if a call comes from an unknown phone number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the generation AI system. The hardware used is a standard server and telephone system for call management. The software used is call recording software and database query software.

[1096] Examples:

[1097] When an elderly user receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to a generative AI system and emotion engine.

[1098] Generative AI System

[1099] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server. A general language model is used as the generative AI model.

[1100] Examples:

[1101] The generative AI system analyzes call content in real time, detecting keywords such as "identity verification," "account," and "funds," while also taking into account the analysis results of the emotion engine to assess fraud risk.

[1102] Example prompt sentence:

[1103] "Assess your risk of fraud based on the following conversation: 'Hello, this is from your bank and I need to verify your account.'"

[1104] Database System

[1105] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the generative AI system deems suspicious to a blacklist in the database, automatically blocking future calls. The hardware used is a standard database server, and the software is a common database management system.

[1106] Examples:

[1107] The server adds the phone number of callers it deems a high risk of fraud to a database and automatically blocks future calls.

[1108] User Notification System

[1109] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. The hardware used is a communication server for sending notifications, and the software used is a notification app or SMS sending system.

[1110] Examples:

[1111] Based on the analysis results of the emotion engine and the risk assessment by the generating AI, a warning message such as "A suspicious call has been detected" is sent via SMS to the elderly person's family.

[1112] Continuous Learning System

[1113] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[1114] Examples:

[1115] Newly detected fraud patterns, along with data from the emotion engine, are saved as training data to update the model of the generative AI system.

[1116] Emotion Engine

[1117] The emotion engine analyzes the user's emotions during a call and sends the results to the generative AI system. It analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc. The hardware uses a computer for voice analysis, and the software uses an emotion analysis algorithm.

[1118] Examples:

[1119] The call content is analyzed, and if the user's emotions indicate anxiety or fear, the results are sent to a generative AI system, which takes this into account when assessing fraud risk.

[1120] Example prompt sentence:

[1121] "Analyze the emotions from this audio data and tell us what emotions the user is feeling."

[1122] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[1123] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1124] Step 1:

[1125] The server detects incoming phone calls and monitors them. If an incoming call comes from an unknown phone number, the server records the call and records details such as the call timestamp, the caller's phone number, and the call length. Specifically, when an elderly user receives a call from an unknown number on their device, the server begins recording the call. The input is the call details, and the output is the recorded audio data.

[1126] Step 2:

[1127] The server queries the caller information of the recorded call in a database system to see if the number is on the blacklist. Specifically, the server queries the caller's phone number in the database and gets the result "not on the blacklist." The input is the caller's phone number, and the output is the database query result.

[1128] Step 3:

[1129] If the server confirms that the caller is not on the blacklist, it records the call and sends the data to the generative AI system and emotion engine. Specifically, after the call is recorded, the server sends the audio data to the generative AI system and emotion engine. The input is the audio data of the recorded call, and the output is the transfer of data to the generative AI system and emotion engine.

[1130] Step 4:

[1131] The generative AI system analyzes the received voice data and detects specific keywords and patterns. The emotion engine also analyzes the voice data and evaluates the user's emotional state. Specifically, the generative AI system detects keywords such as "identity verification," "account," and "funds" from the call, and the emotion engine analyzes the tone and speed of the user's voice to determine whether they are feeling anxious. The input is the voice data of the call content, and the output is specific keywords and the user's emotional evaluation result.

[1132] Step 5:

[1133] The server evaluates whether the risk of fraud is high based on the fraud risk assessment results provided by the generative AI system and the analysis results of the emotion engine. Specifically, if the generative AI system returns a high risk score and the emotion engine concludes that the user is anxious, the server determines that the risk of fraud is high. The inputs are the assessment results of the generative AI system and the analysis results of the emotion engine, and the output is the risk assessment result.

[1134] Step 6:

[1135] If the server determines that there is a high risk of fraud, it sends the generated auto-answer message to the caller and disconnects the call.Specifically, a message such as "There is no one in charge at the moment. We will contact you later" is played and the call is disconnected.The input is a template for the auto-answer message, and the output is sending a message to the caller and disconnecting the call.

[1136] Step 7:

[1137] The server adds the caller's phone number to a blacklist in the database system. Specifically, the server registers the caller's phone number in the database, and future calls are automatically blocked. The input is the caller's phone number, and the output is the registration result in the database.

[1138] Step 8:

[1139] The user notification system notifies users and their families in real time about detected fraud risks. Specifically, it sends a warning message to the elderly family member's smartphone via SMS saying, "A suspicious call has been detected." The input is the fraud risk assessment result, and the output is the sending of a notification message.

[1140] Step 9:

[1141] The continuous learning system saves newly detected fraud patterns and emotion data as learning data and periodically updates the generative AI system and emotion engine models. Specifically, the detected fraud patterns and emotion data are saved in a database, and the model is periodically updated. The input is the newly detected fraud patterns and emotion data, and the output is the updated AI model.

[1142] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[1143] (Application example 2)

[1144] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1145] In recent years, the number of elderly people who fall victim to special frauds over the phone has been increasing. However, conventional fraud prevention systems are finding it difficult to respond effectively to the increasing diversity and sophistication of fraud methods. In addition, since it is difficult for elderly people to judge the content of phone calls themselves, there is a need for a system that can prevent fraud damage before it occurs. In particular, the development of an advanced fraud prevention system that automatically responds to fraud methods and combines real-time emotion analysis is a challenge.

[1146] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automatic response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; emotion analysis means for evaluating the user's emotional state; and means for complementing the fraud risk assessment based on the user's emotional state. This significantly reduces the risk of elderly people falling victim to special telephone frauds and provides an environment in which they can live safely.

[1147] "Means for receiving calls and checking the caller information against a database" is a function that automatically obtains information about an incoming call and compares the caller information with a pre-registered database.

[1148] "Means for automatically blocking calls when the caller is a suspicious phone number" is a function that automatically blocks incoming calls from suspicious phone numbers based on the results of a database match.

[1149] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of the conversation in real time during a call and assesses the possibility of fraud based on that content.

[1150] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" is a function that automatically generates a response message and sends it to the caller when the system determines that there is a high risk of fraud.

[1151] "Means for storing and managing blacklisted telephone numbers in a database" refers to a management function that stores telephone numbers of callers deemed suspicious in a database and consistently blocks subsequent calls.

[1152] "Means of notifying users and their families in real time when calls that pose a high risk of fraud are detected" is a function that immediately sends a warning to users and their families when calls that pose a high risk of fraud are detected.

[1153] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that adds newly discovered fraud methods and patterns as learning data and improves the accuracy of the analysis model.

[1154] The "emotion analysis means for evaluating the user's emotional state" is a function that analyzes the user's tone of voice and choice of words during a call and evaluates their emotional state.

[1155] "Means to complement fraud risk assessment based on emotional state" is a function that further refines fraud risk assessment based on the results of user emotional analysis.

[1156] To implement the present invention, a system including the following procedures and components is used.

[1157] System Overview

[1158] The system consists of a server that receives and analyzes calls in real time, a terminal that monitors the user's call status, and a notification system that notifies users and their families based on the results of the call analysis.

[1159] 1. Receiving calls and querying caller information

[1160] The server automatically records calls received by the user and checks the caller information against a database that includes a blacklist of suspicious phone numbers.

[1161] 2. Real-time analysis and automatic response

[1162] The server analyzes the call content in real time and assesses the risk of fraud based on the content of the conversation and the keywords used.

[1163] If the risk of fraud is determined to be high, the server generates an automated response message and sends it to the caller, designed to prevent fraud before it occurs.

[1164] 3. Emotion analysis

[1165] The server uses an emotion engine to analyze the tone, rate and emotional fluctuations of the user's voice during the call to assess the user's emotional state.

[1166] Sentiment analysis results are used to complement fraud risk assessment.

[1167] 4. Notification System

[1168] If a fraudulent call is detected, the notification system will send a real-time notification to the user and their family via SMS, notification app, or email.

[1169] 5. Blacklist Management

[1170] The server stores phone numbers that are deemed suspicious in a blacklist in its database and automatically blocks future calls.

[1171] 6. Continuous Learning System

[1172] The continuous learning system saves newly detected fraud patterns as training data to improve the accuracy of the analysis model.

[1173] Hardware and software used

[1174] 1. Hardware

[1175] Server: Data processing device for analysis and notification

[1176] Smartphone: Records user calls and sends them to a server

[1177] 2. Software

[1178] Call reception app: Records user calls and sends the data to a server

[1179] Generative AI system: Analyzes call content and assesses fraud risk

[1180] Emotion Engine: Analyzes the user's emotional state

[1181] Notification app: Sends notifications about fraud risks

[1182] Specific examples

[1183] 1. Receiving and analyzing calls

[1184] When a user receives a call from an unknown number, the server records the call and checks the caller's number against a database. If the call is not blacklisted, the server analyzes the call in real time.

[1185] 2. Sentiment Analysis and Risk Assessment

[1186] The server uses an emotion engine to analyze the user's emotional state, and the generative AI system analyzes keywords in the call content. The results of the emotion analysis and the call content analysis are combined to assess the fraud risk.

[1187] 3. Auto-replies and notifications

[1188] If the risk of fraud is deemed high, the server generates an automated response message and sends it to the caller, while simultaneously sending real-time notifications to the user and their family members via SMS, notification app, or email.

[1189] Prompt Sentence Examples

[1190] Call content: {Call text}

[1191] Emotional state: {User emotion}

[1192] Criteria: Assess the fraud risk and return a "High Fraud Risk" message if the risk is high.

[1193] In this way, this system reduces the risk of users becoming victims of special fraud and provides an environment in which users can make calls with peace of mind.

[1194] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1195] Step 1:

[1196] When a user receives a call, the device records the call and obtains the caller's phone number. The obtained call data and caller information are sent to the server. The input is the incoming call's voice data and the caller's phone number, and the output is that the data is sent to the server.

[1197] Step 2:

[1198] The server checks the received caller information against the database. If the check finds that the caller is registered on the blacklist, the server automatically blocks the call. The input is the caller's phone number, and the output is the check result and the call blocking process.

[1199] Step 3:

[1200] If the caller is not on the blacklist, the server analyzes the call in real time using a generative AI system that performs keyword analysis and pattern matching on the call content. The input is the audio data of the call, and the output is the fraud risk assessment result.

[1201] Step 4:

[1202] The server uses an emotion engine to analyze the tone, speed, and emotional fluctuations of the user's voice during the call to evaluate the user's emotional state. The input is the voice data of the call, and the output is the evaluation result of the user's emotional state.

[1203] Step 5:

[1204] The server evaluates the risk of fraud by combining the results of keyword analysis by the generative AI system and the results of the emotional state evaluation by the emotion engine. If the risk of fraud is determined to be high, the server generates an automatic response message and sends it to the caller. The input is the analysis results of the generative AI system and the evaluation results of the emotion engine, and the output is the automatic response message.

[1205] Step 6:

[1206] If a fraud risk is detected, the server sends a notification to the user and their family members via SMS, a notification app, or email. The input is the fraud risk assessment result, and the output is the notification message.

[1207] Step 7:

[1208] The server adds any phone numbers it deems suspicious to a blacklist in its database, automatically blocking future calls. The input is the suspicious phone number, and the output is the registration status in the database.

[1209] Step 8:

[1210] The continuous learning system improves accuracy by saving newly detected fraud patterns as training data and updating the analytical models of the generative AI system and emotion engine. The input is the newly detected fraud patterns, and the output is the updated models.

[1211] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1212] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1213] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1214] [Fourth embodiment]

[1215] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1216] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1217] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1218] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1219] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1220] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1221] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1222] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1223] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1224] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1225] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1226] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1227] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1228] The present invention is an AI telephone handling service system for helping elderly people avoid falling victim to special fraud, and can be implemented as follows.

[1229] System Overview

[1230] The system consists of the following main components:

[1231] 1. Call Reception Server

[1232] 2. Generative AI Systems

[1233] 3. Database System

[1234] 4. User Notification System

[1235] 5. Continuous Learning System

[1236] Program processing

[1237] Call acceptance server

[1238] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[1239] Generative AI System

[1240] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[1241] Database System

[1242] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[1243] User Notification System

[1244] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[1245] Continuous Learning System

[1246] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[1247] Specific examples

[1248] Receiving and analyzing calls

[1249] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the database for the caller's phone number. If the call is not blacklisted, the server sends the call to the AI ​​system.

[1250] Analysis and automatic response by generative AI

[1251] The AI ​​system analyzes the content of calls in real time, detecting keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, the AI ​​generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[1252] Call blocking and notifications

[1253] The server plays the automated response message sent by the AI ​​system to the caller and disconnects the call. At the same time, the target phone number is added to the blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[1254] Continuous learning and updates

[1255] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[1256] In this way, this system can protect the elderly from the risk of special fraud and provide an environment where they can live in peace of mind.

[1257] The processing flow will be explained below.

[1258] Step 1:

[1259] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[1260] Step 2:

[1261] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[1262] Step 3:

[1263] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[1264] Step 4:

[1265] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generation AI system.

[1266] Step 5:

[1267] The generative AI system analyzes incoming call data to assess fraud risk, detecting keywords and conversation patterns such as "identity verification," "account," and "funds."

[1268] Step 6:

[1269] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[1270] Step 7:

[1271] The server plays the automated response message sent by the generating AI system to the caller and then disconnects the call.

[1272] Step 8:

[1273] The server adds the caller's phone number to a blacklist in its database, automatically blocking future calls.

[1274] Step 9:

[1275] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected via SMS, notification app, or email.

[1276] Step 10:

[1277] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the generative AI system's model, thereby improving the analysis accuracy of the entire system.

[1278] Example 1

[1279] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1280] In modern society, the risk of elderly people becoming victims of special frauds is increasing. Such frauds are becoming more sophisticated, with a particularly rapid increase in fraud methods over the phone. Elderly people have difficulty recognizing fraud and are often more susceptible to falling victim. Therefore, there is a need for a system that can automatically assess fraud risks and respond quickly. Furthermore, there is a need for a means to notify elderly people and their families of fraud risks in real time, and a system that can continuously respond to new fraud patterns is also needed.

[1281] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1282] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means for recording the content of calls and sending it to a generative AI model system for analysis; and means for generating an automated response message based on the analysis results and sending it to the server. This protects elderly people from the risk of special fraud and allows them to receive prompt fraud warnings. It also enables continuous learning and improvement of the accuracy of the analysis model to respond to new fraud techniques.

[1283] "Means for receiving calls and checking the database for caller information" refers to a function that accepts external telephone calls and checks the caller's telephone number against an internal database to see if the caller matches the registered information.

[1284] "Means for automatically blocking calls when the caller is a suspicious phone number" refers to a function that automatically disconnects the call and prevents the call from being forwarded if the caller's phone number is blacklisted as a result of a database check.

[1285] "A means of analyzing the content of calls in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that uses voice recognition to instantly analyze the content of calls and determine whether there is a possibility of fraud if the call is not on the blacklist.

[1286] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" refers to a function that automatically generates a standard response message and sends it to the caller when the analysis of the content of the call indicates a high risk of fraud.

[1287] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for recording telephone numbers of suspicious callers in a database and continuously managing and updating that information.

[1288] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that sends an immediate warning notification to users and their families when a call that poses a high risk of fraud is detected.

[1289] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function for improving the accuracy of fraud detection by accumulating newly detected fraud methods and patterns as learning data and periodically updating the analysis algorithm model.

[1290] "Means for recording the contents of a call and transmitting it to a generative AI model system for analysis" refers to a function that records the audio during a call and transmits the recorded data to an analysis system that uses a generative AI model, thereby performing processing to analyze the contents of the call.

[1291] "Means for generating an automatic response message based on the analysis results and sending it to the server" refers to a function that creates an appropriate automatic response message based on the results of analysis by the generative AI model system and sends that message to the server.

[1292] MODE FOR CARRYING OUT THE INVENTION

[1293] This invention is an AI telephone handling service system designed to protect elderly people from falling victim to special fraud. This system consists of the following main components: a call reception server, a generation AI system, a database system, a user notification system, and a continuous learning system. Below, we provide a detailed explanation of each component and explain how the system works.

[1294] System Overview

[1295] Call acceptance server

[1296] The call reception server monitors all incoming calls. When a new call comes in, it captures the call information and checks whether it is an unknown phone number. The server then queries the caller's phone number in a database system to determine whether the number is blacklisted. If the caller is not blacklisted, it records the call and sends the data to the generation AI system.

[1297] Examples of hardware and software used: VoIP devices, communication control servers

[1298] Generative AI System

[1299] The generative AI system analyzes incoming call data and assesses the risk of fraud. The system includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an automated response message and sends it to the server.

[1300] Examples of hardware and software used: Natural Language Processing (NLP) engine, Python, TensorFlow library

[1301] Database System

[1302] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the AI ​​system determines to be suspicious to a blacklist in the database, automatically blocking future calls.

[1303] Examples of hardware and software used: SQL database server

[1304] User Notification System

[1305] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[1306] Examples of hardware and software used: notification server, SMS gateway, notification application

[1307] Continuous Learning System

[1308] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[1309] Examples of hardware and software used: machine learning platform, data storage system

[1310] Specific examples

[1311] Receiving and analyzing calls

[1312] For example, if an elderly person receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to the AI ​​system.

[1313] Analysis and automatic response by generative AI

[1314] The AI ​​generation system analyzes the content of calls in real time. For example, it detects keywords such as "identity verification," "account," and "funds." If it determines there is a high risk of fraud, it generates an automatic response message such as "There is no one in charge at the moment. We will contact you later," and sends it to the server.

[1315] Call blocking and notifications

[1316] The server plays the automated message sent by the AI ​​system to the caller and disconnects the call. At the same time, the server adds the phone number to a blacklist. In addition, the user notification system sends a warning message to the elderly person's family saying, "A suspicious call has been detected."

[1317] Continuous learning and updates

[1318] The continuous learning system saves detected fraud patterns as training data and periodically updates the generative AI's language model, improving fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[1319] Prompt Sentence Examples

[1320] "Analyze the content of this call, and if suspicious keywords (such as 'identity verification,' 'account,' 'funds,' etc.) are detected, determine that it may be a scam and generate an automated response message."

[1321] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1322] Step 1:

[1323] The server monitors all incoming calls. When a new call is received, it captures the call information (caller's phone number, date and time, etc.). (Input): New incoming call. (Data processing): Capture and record call information. (Output): Caller information.

[1324] Step 2:

[1325] The server queries the caller's phone number in a database system. (Input): Caller information for incoming calls. (Data calculation): Query the database. (Output): Query results (blacklist match).

[1326] Step 3:

[1327] The server judges the query result and checks whether the caller is on the blacklist. If the caller is on the blacklist, the server will automatically block the call. (Input): Database query result. (Data calculation): Judgment process. (Output): Call blocking.

[1328] Step 4:

[1329] If the caller is not on the blacklist, the server records the call and sends the data to the generation AI system. (Input): Call information that does not fall under the blacklist. (Data processing): Recording of call content and data transmission. (Output): Recorded data.

[1330] Step 5:

[1331] The generative AI system analyzes the received call data and assesses the risk of fraud. (Input): Recorded data. (Data calculation): Analysis of call content and risk assessment using natural language processing. (Output): Risk assessment results (presence or absence of fraud risk and its degree).

[1332] Step 6:

[1333] If the risk of fraud is determined to be high, the AI ​​system generates an automatic response message and sends it to the server. (Input): Risk assessment result. (Data calculation): Generation of automatic response message. (Output): Automatic response message.

[1334] Step 7:

[1335] The server plays the automated response message sent by the AI ​​generation system to the caller and disconnects the call. At the same time, it adds the target phone number to the blacklist. (Input): Auto-response message, caller information. (Data processing): Plays the message and disconnects the call, updates the blacklist. (Output): Phone number added to the blacklist.

[1336] Step 8:

[1337] The user notification system sends real-time notifications to users and their families when fraudulent calls are detected. (Input): Risk assessment results and blacklist update information. (Data calculation): Notification message generation. (Output): Notification message (SMS, notification app, email).

[1338] Step 9:

[1339] The continuous learning system saves newly detected fraud patterns as training data and periodically updates the model of the generative AI system. (Input): Newly detected fraud data. (Data calculation): Update training data and retrain the model. (Output): Improved AI model.

[1340] (Application example 1)

[1341] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1342] With the elderly currently at a higher risk of falling victim to special frauds, there is a need for a system to protect them from telephone fraud. However, existing systems have difficulty analyzing call content in real time and immediately assessing the risk of fraud. Even if calls with a high risk of fraud are detected, appropriate measures may not be taken promptly, making them insufficient to ensure the safety of the elderly. Furthermore, there is a risk that responses to new fraud methods will be delayed, reducing the accuracy of the system. It is necessary to solve these issues and provide an environment in which the elderly can live with peace of mind.

[1343] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1344] In this invention, the server includes: means for receiving calls and querying a database for caller information; means for automatically blocking calls if the caller is a suspicious phone number; means for analyzing the content of calls in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automated response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying users and their families in real time when a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; means operated via a mobile communication terminal with a call monitoring function; and means with a security function for sending the generated automated response message to a designated recipient. This safely protects elderly people from special frauds, analyzes suspicious calls in real time to take appropriate measures, and enables elderly people and their families to quickly understand the risks.

[1345] A "mobile communication terminal with call monitoring function" refers to a portable communication device that has the function of receiving calls and monitoring their content.

[1346] The "means for querying a database for caller information" refers to a system that has the function of querying and confirming caller information from a database that stores information about the caller.

[1347] "Automatic call blocking" is a feature that automatically blocks calls when a suspicious phone number or call is detected.

[1348] "Means for assessing the risk of fraud" refers to a function that analyzes the content of a call and determines the likelihood that the call is fraudulent.

[1349] "Means for generating and sending automatic response messages to callers" refers to a function that automatically creates a pre-defined message and sends it to callers when it is determined that there is a high risk of fraud.

[1350] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function that adds telephone numbers that are determined to be at high risk of fraud to a specific list and manages that list in a database.

[1351] "Means of notifying users and their families in real time" is a function that quickly notifies users and their families of any calls that pose a risk of fraud when such calls are detected.

[1352] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" refers to a function that allows the system to learn newly discovered fraud methods and patterns, update the analysis model based on them, and improve accuracy.

[1353] The present invention relates to a system for protecting elderly people from special frauds, which is implemented using a mobile communication terminal and a server. Specific embodiments of the system are described below.

[1354] 1. System Configuration

[1355] The system consists of the following main components:

[1356] 1. Mobile communication terminals

[1357] 2. Call Reception Server

[1358] 3. Generative AI Systems

[1359] 4. Database Systems

[1360] 5. User Notification System

[1361] 6. Continuous Learning System

[1362] 2. System Operation

[1363] Mobile communication terminal

[1364] A mobile communication terminal is a portable communication device equipped with a call monitoring function. When a call is received, the terminal records the call and transmits the caller information to a server.

[1365] Call acceptance server

[1366] The server receives calls and checks the caller's phone number against a database. If a call comes from an unknown number, it records the call and sends it to a generative AI system. If the number is suspicious, the call is automatically blocked.

[1367] Generative AI System

[1368] The generative AI system analyzes the content of calls in real time and assesses the risk of fraud. The AI ​​model converts the voice data into text and analyzes the text to determine the risk of fraud. If it determines that the risk of fraud is high, it generates an appropriate automated response message and sends it to the server.

[1369] Database System

[1370] The database system stores and manages information on suspicious phone numbers and fraudulent methods. For example, a blacklist is managed in this database, and the server adds new blacklist items based on information from the generative AI system.

[1371] User Notification System

[1372] The user notification system will send real-time notifications to users and their families when a fraudulent call is detected. This notification will be sent via SMS, a notification app, or email, allowing users and their families to quickly understand the risk.

[1373] Continuous Learning System

[1374] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[1375] 3. Hardware and Software

[1376] The following hardware and software are used to implement this system:

[1377] Hardware: Smartphones, servers

[1378] Software: Python, Requests library, smtplib, custom API (call acceptance, generative AI, database management)

[1379] 4. Examples and prompts

[1380] Specific examples

[1381] An elderly person installs a smartphone app. One day, the elderly person receives a call from an unknown number. The app monitors the call and determines it to be suspicious. A notification is sent to the family saying, "There was a suspicious call from 0123456789."

[1382] Prompt example

[1383] "Please explain the functionality of an AI application that analyzes calls from unknown numbers in real time to protect seniors from specialized fraud."

[1384] This will ensure that elderly people are safely protected from special fraud, suspicious calls will be analyzed in real time and appropriate measures will be taken, and elderly people and their families will be able to quickly understand the risks.

[1385] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1386] Step 1:

[1387] The server receives calls from mobile communication terminals, obtains the caller's phone number, and checks the caller information against a database to see if the phone number is suspicious.

[1388] Specifically, it receives the caller's phone number as input, searches for the corresponding phone number in the database, and returns a flag indicating whether the caller's phone number is suspicious.

[1389] Step 2:

[1390] If the server determines that the caller is a suspicious number, it will automatically block the call.

[1391] Specifically, it receives a suspicious phone number flag as input, blocks the call, and returns a confirmation message that the call has been blocked as output.

[1392] Step 3:

[1393] If the caller is not a suspicious phone number, the server records the call and sends the data to the generation AI system.

[1394] Specifically, it receives voice call data as input, sends it to the AI ​​analysis system, and returns a message confirming the transmission of the voice call data as output.

[1395] Step 4:

[1396] The generative AI system analyzes incoming call content in real time and assesses the risk of fraud.

[1397] Specifically, the system receives voice call data as input, performs text analysis using an AI model, and returns the fraud risk assessment results to the server as output.

[1398] Step 5:

[1399] The server receives the evaluation results from the generative AI system and, if it determines that there is a high risk of fraud, generates an automated response message to send to the caller.

[1400] Specifically, it receives the fraud risk assessment result as input, generates an automatic response message, and returns the response message to be sent to the caller as output.

[1401] Step 6:

[1402] The server adds phone numbers that are deemed to pose a fraud risk to a blacklist in its database, automatically blocking future calls.

[1403] Specifically, it receives a fraudulent phone number as input, registers it in a blacklist in the database, and returns a message confirming the registration to the blacklist as output.

[1404] Step 7:

[1405] The user notification system notifies users and their families in real time when calls that pose a risk of fraud are detected.

[1406] It takes the fraud risk assessment result as input, generates notification data, and sends notifications via SMS, notification app, or email as output.

[1407] Step 8:

[1408] The continuous learning system stores newly detected fraud patterns as training data and periodically updates the generative AI system's model.

[1409] Specifically, it receives new fraud pattern data as input, stores it in the training database, and returns a confirmation message for the updated AI model as output.

[1410] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1411] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds, and in particular, improves the accuracy of fraud detection by combining it with an emotion engine that recognizes the user's emotions. The following describes specific embodiments of the present invention.

[1412] System Overview

[1413] The system consists of the following main components:

[1414] 1. Call Reception Server

[1415] 2. Generative AI Systems

[1416] 3. Database System

[1417] 4. User Notification System

[1418] 5. Continuous Learning System

[1419] 6. Emotion Engine

[1420] Program processing

[1421] Call acceptance server

[1422] The server monitors all incoming calls, and if a call comes from an unknown number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the AI ​​generation system.

[1423] Generative AI System

[1424] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[1425] Database System

[1426] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that are judged to be suspicious by the AI ​​generation system to a blacklist in the database, automatically blocking future calls.

[1427] User Notification System

[1428] The user notification system will send real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. This allows users and their families to quickly learn about fraud risks.

[1429] Continuous Learning System

[1430] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[1431] Emotion Engine

[1432] The emotion engine analyzes the user's emotions during the call and sends the results to the generative AI system. The emotion engine analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc.

[1433] Specific examples

[1434] Receiving and analyzing calls

[1435] For example, if a call comes in to a senior citizen's phone from an unknown number, the server will record the call and query the caller's phone number against a database. If the database does not match any blacklist, the call will be sent to the generative AI system and emotion engine.

[1436] Collaboration between analysis and generative AI using an emotion engine

[1437] The generative AI system analyzes the call content in real time, detecting keywords such as "identity verification," "account," and "funds." At the same time, the emotion engine analyzes the voice and evaluates the user's emotional state. If the emotion engine determines that the user's emotions are abnormal, such as anxiety or fear, the generative AI will take this into account when assessing the fraud risk.

[1438] Auto-response and blacklist management

[1439] If the AI ​​system determines that there is a high risk of fraud, it generates an automated response message such as "We are currently unavailable. We will contact you shortly" and sends it to the server. The server plays this message to the caller and disconnects the call. The server then adds the caller's phone number to a blacklist in its database.

[1440] Notification and follow-up

[1441] Based on the analysis results of the emotion engine and the risk assessment of the generative AI, the user notification system will send warning messages such as "A suspicious call has been detected" to the elderly person's family members, and will also provide further detailed advice and follow-up notifications if necessary.

[1442] Continuous learning and updates

[1443] The continuous learning system saves the detected fraud patterns and data from the emotion engine as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving the fraud detection accuracy from the next time onwards and ensuring the safety of the elderly.

[1444] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[1445] The processing flow will be explained below.

[1446] Step 1:

[1447] The server monitors all incoming calls and, if any call is received, captures the caller's phone number.

[1448] Step 2:

[1449] The server checks the caller's phone number against a database to see if the caller is on a blacklist or whitelist.

[1450] Step 3:

[1451] If the caller is on the blacklist, the server automatically blocks the call and plays an automated message to the caller, such as "There is no one in the office right now. We will contact you shortly."

[1452] Step 4:

[1453] If the caller is not on the whitelist, the server records the call in real time and sends the data to the generative AI system and emotion engine.

[1454] Step 5:

[1455] The generative AI system analyzes the transmitted call data and assesses the risk of specialized fraud. The analysis involves detecting keywords and conversation patterns, such as "identity verification," "account," and "funds."

[1456] Step 6:

[1457] The emotion engine analyzes voice data during a call to detect the user's tone and emotional fluctuations, thereby assessing whether the user is experiencing emotions such as anxiety, fear, or anger.

[1458] Step 7:

[1459] The generative AI system adjusts the fraud risk assessment based on the sentiment analysis results from the emotion engine. For example, if the emotion engine detects anxiety or fear in the user, it will determine that the fraud risk is high.

[1460] Step 8:

[1461] If the generative AI system determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server.

[1462] Step 9:

[1463] The server plays the automated response message sent by the AI ​​system to the caller, disconnects the call, and adds the caller's phone number to a blacklist in the database.

[1464] Step 10:

[1465] The user notification system will send real-time notifications to users and their families based on the analysis results of the emotion engine and the risk assessment of the generative AI system via SMS, notification app, or email.

[1466] Step 11:

[1467] The notification message will include something like, "A suspicious call has been detected. Please contact your system administrator for more information." In some cases, more detailed advice or follow-up notifications will be provided.

[1468] Step 12:

[1469] The continuous learning system saves detected fraud patterns and emotion engine data as training data and periodically updates the generative AI's language model and emotion analysis model, thereby improving fraud detection accuracy in future iterations.

[1470] Example 2

[1471] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1472] There is a problem that exposes elderly people to the risk of special fraud. Existing telephone systems may not be able to detect fraud risks, increasing the likelihood that elderly people will become victims of fraud. In addition, it is difficult to quickly notify and respond to fraud risks, so prompt measures are needed to minimize damage.

[1473] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for analyzing the user's emotions and reflecting them in the fraud risk assessment; means for generating and sending an automatic response message to the caller if the fraud risk is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call with a risk of fraud is detected; and means for learning newly detected fraud patterns and improving the accuracy of the analysis model. This protects elderly people from fraud, enables rapid and accurate detection of fraud risks, and enables appropriate responses.

[1474] "Means for receiving calls and checking the database for caller information" is a function that detects an incoming call and checks the caller's telephone number against a database to confirm it.

[1475] "Means to automatically block calls if the caller is a suspicious phone number" is a function that automatically prevents calls from being received if the caller's phone number is included in a blacklist in the database.

[1476] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of a conversation during a call in real time and determines the risk of fraud based on that content.

[1477] "Means for analyzing user emotions and reflecting them in fraud risk assessment" is a function that analyzes changes in the user's tone of voice and emotions during a call and uses the results to help assess fraud risk.

[1478] "Means for generating and sending an automatic response message to a caller when a high risk of fraud is determined" is a function that sends an automatically created message to a caller for calls that are assessed as having a high risk of fraud.

[1479] "Means for storing and managing blacklisted telephone numbers in a database" refers to a function for registering telephone numbers that are determined to pose a risk of fraud in a database and for retaining and managing that information.

[1480] "Means of notifying users and their families in real time when a call that poses a high risk of fraud is detected" is a function that notifies users and their families in real time when a call that poses a high risk of fraud occurs.

[1481] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that learns newly discovered fraud methods and patterns and improves the accuracy of the fraud detection model for the entire system.

[1482] The present invention is an AI telephone handling service system that helps elderly people avoid falling victim to special frauds. In particular, by combining it with an emotion engine that recognizes the user's emotions, the accuracy of fraud detection is improved. This system is realized using hardware and software. The following describes in detail an embodiment of the present invention.

[1483] System Overview

[1484] The system consists of the following main components:

[1485] 1. Call Reception Server

[1486] 2. Generative AI Systems

[1487] 3. Database System

[1488] 4. User Notification System

[1489] 5. Continuous Learning System

[1490] 6. Emotion Engine

[1491] Call acceptance server

[1492] The server monitors all incoming calls, and if a call comes from an unknown phone number, it temporarily records the call and collects caller information. It then checks the caller's phone number against a database to see if it is on a blacklist. If the caller is not on the blacklist, it records the call and sends the data to the generation AI system. The hardware used is a standard server and telephone system for call management. The software used is call recording software and database query software.

[1493] Examples:

[1494] When an elderly user receives a call from an unknown number, the server records the call and queries the caller's phone number against a database. If the call is not blacklisted, the call is sent to a generative AI system and emotion engine.

[1495] Generative AI System

[1496] The generative AI system analyzes the received call data and assesses the risk of special fraud. The AI ​​model includes a language model and analyzes keywords and patterns in the call content. If it determines that there is a high risk of fraud, it generates an appropriate automated response message and sends it to the server. A general language model is used as the generative AI model.

[1497] Examples:

[1498] The generative AI system analyzes call content in real time, detecting keywords such as "identity verification," "account," and "funds," while also taking into account the analysis results of the emotion engine to assess fraud risk.

[1499] Example prompt sentence:

[1500] "Assess your risk of fraud based on the following conversation: 'Hello, this is from your bank and I need to verify your account.'"

[1501] Database System

[1502] The database system stores and manages information on suspicious phone numbers and fraudulent methods. The server adds phone numbers that the generative AI system deems suspicious to a blacklist in the database, automatically blocking future calls. The hardware used is a standard database server, and the software is a common database management system.

[1503] Examples:

[1504] The server adds the phone number of callers it deems a high risk of fraud to a database and automatically blocks future calls.

[1505] User Notification System

[1506] The user notification system sends real-time notifications to users and their families when high-risk fraud calls are detected. Notifications can be sent via SMS, a notification app, or email. The hardware used is a communication server for sending notifications, and the software used is a notification app or SMS sending system.

[1507] Examples:

[1508] Based on the analysis results of the emotion engine and the risk assessment by the generating AI, a warning message such as "A suspicious call has been detected" is sent via SMS to the elderly person's family.

[1509] Continuous Learning System

[1510] The continuous learning system saves newly detected fraud patterns as learning data and periodically updates the generative AI system's model, improving the system's overall analysis accuracy and enabling it to adapt to the latest fraud techniques.

[1511] Examples:

[1512] Newly detected fraud patterns, along with data from the emotion engine, are saved as training data to update the model of the generative AI system.

[1513] Emotion Engine

[1514] The emotion engine analyzes the user's emotions during a call and sends the results to the generative AI system. It analyzes the tone, speed, and emotional fluctuations of the voice to detect whether the user is feeling fear, confusion, anger, etc. The hardware uses a computer for voice analysis, and the software uses an emotion analysis algorithm.

[1515] Examples:

[1516] The call content is analyzed, and if the user's emotions indicate anxiety or fear, the results are sent to a generative AI system, which takes this into account when assessing fraud risk.

[1517] Example prompt sentence:

[1518] "Analyze the emotions from this audio data and tell us what emotions the user is feeling."

[1519] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[1520] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1521] Step 1:

[1522] The server detects incoming phone calls and monitors them. If an incoming call comes from an unknown phone number, the server records the call and records details such as the call timestamp, the caller's phone number, and the call length. Specifically, when an elderly user receives a call from an unknown number on their device, the server begins recording the call. The input is the call details, and the output is the recorded audio data.

[1523] Step 2:

[1524] The server queries the caller information of the recorded call in a database system to see if the number is on the blacklist. Specifically, the server queries the caller's phone number in the database and gets the result "not on the blacklist." The input is the caller's phone number, and the output is the database query result.

[1525] Step 3:

[1526] If the server confirms that the caller is not on the blacklist, it records the call and sends the data to the generative AI system and emotion engine. Specifically, after the call is recorded, the server sends the audio data to the generative AI system and emotion engine. The input is the audio data of the recorded call, and the output is the transfer of data to the generative AI system and emotion engine.

[1527] Step 4:

[1528] The generative AI system analyzes the received voice data and detects specific keywords and patterns. The emotion engine also analyzes the voice data and evaluates the user's emotional state. Specifically, the generative AI system detects keywords such as "identity verification," "account," and "funds" from the call, and the emotion engine analyzes the tone and speed of the user's voice to determine whether they are feeling anxious. The input is the voice data of the call content, and the output is specific keywords and the user's emotional evaluation result.

[1529] Step 5:

[1530] The server evaluates whether the risk of fraud is high based on the fraud risk assessment results provided by the generative AI system and the analysis results of the emotion engine. Specifically, if the generative AI system returns a high risk score and the emotion engine concludes that the user is anxious, the server determines that the risk of fraud is high. The inputs are the assessment results of the generative AI system and the analysis results of the emotion engine, and the output is the risk assessment result.

[1531] Step 6:

[1532] If the server determines that there is a high risk of fraud, it sends the generated auto-answer message to the caller and disconnects the call.Specifically, a message such as "There is no one in charge at the moment. We will contact you later" is played and the call is disconnected.The input is a template for the auto-answer message, and the output is sending a message to the caller and disconnecting the call.

[1533] Step 7:

[1534] The server adds the caller's phone number to a blacklist in the database system. Specifically, the server registers the caller's phone number in the database, and future calls are automatically blocked. The input is the caller's phone number, and the output is the registration result in the database.

[1535] Step 8:

[1536] The user notification system notifies users and their families in real time about detected fraud risks. Specifically, it sends a warning message to the elderly family member's smartphone via SMS saying, "A suspicious call has been detected." The input is the fraud risk assessment result, and the output is the sending of a notification message.

[1537] Step 9:

[1538] The continuous learning system saves newly detected fraud patterns and emotion data as learning data and periodically updates the generative AI system and emotion engine models. Specifically, the detected fraud patterns and emotion data are saved in a database, and the model is periodically updated. The input is the newly detected fraud patterns and emotion data, and the output is the updated AI model.

[1539] In this way, this system can protect elderly people from the risk of special fraud and provide an environment where they can live with peace of mind.The combination of an emotion engine enables advanced fraud detection that takes into account the user's emotional state.

[1540] (Application example 2)

[1541] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1542] In recent years, the number of elderly people who fall victim to special frauds over the phone has been increasing. However, conventional fraud prevention systems are finding it difficult to respond effectively to the increasing diversity and sophistication of fraud methods. In addition, since it is difficult for elderly people to judge the content of phone calls themselves, there is a need for a system that can prevent fraud damage before it occurs. In particular, the development of an advanced fraud prevention system that automatically responds to fraud methods and combines real-time emotion analysis is a challenge.

[1543] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving a call and querying a database for caller information; means for automatically blocking the call if the caller is a suspicious phone number; means for analyzing the content of the call in real time and assessing the risk of fraud if the caller is not a suspicious phone number; means for generating and sending an automatic response message to the caller if the risk of fraud is determined to be high; means for storing and managing blacklisted phone numbers in a database; means for notifying the user and their family in real time if a call posing a risk of fraud is detected; means for learning newly detected fraud patterns and improving the accuracy of the analysis model; emotion analysis means for evaluating the user's emotional state; and means for complementing the fraud risk assessment based on the user's emotional state. This significantly reduces the risk of elderly people falling victim to special telephone frauds and provides an environment in which they can live safely.

[1544] "Means for receiving calls and checking the caller information against a database" is a function that automatically obtains information about an incoming call and compares the caller information with a pre-registered database.

[1545] "Means for automatically blocking calls when the caller is a suspicious phone number" is a function that automatically blocks incoming calls from suspicious phone numbers based on the results of a database match.

[1546] "A means of analyzing the content of a call in real time and assessing the risk of fraud when the caller is not a suspicious phone number" is a function that analyzes the content of the conversation in real time during a call and assesses the possibility of fraud based on that content.

[1547] "Means for generating and sending an automatic response message to the caller when it is determined that there is a high risk of fraud" is a function that automatically generates a response message and sends it to the caller when the system determines that there is a high risk of fraud.

[1548] "Means for storing and managing blacklisted telephone numbers in a database" refers to a management function that stores telephone numbers of callers deemed suspicious in a database and consistently blocks subsequent calls.

[1549] "Means of notifying users and their families in real time when calls that pose a high risk of fraud are detected" is a function that immediately sends a warning to users and their families when calls that pose a high risk of fraud are detected.

[1550] "Means for learning newly detected fraud patterns and improving the accuracy of the analysis model" is a function that adds newly discovered fraud methods and patterns as learning data and improves the accuracy of the analysis model.

[1551] The "emotion analysis means for evaluating the user's emotional state" is a function that analyzes the user's tone of voice and choice of words during a call and evaluates their emotional state.

[1552] "Means to complement fraud risk assessment based on emotional state" is a function that further refines fraud risk assessment based on the results of user emotional analysis.

[1553] To implement the present invention, a system including the following procedures and components is used.

[1554] System Overview

[1555] The system consists of a server that receives and analyzes calls in real time, a terminal that monitors the user's call status, and a notification system that notifies users and their families based on the results of the call analysis.

[1556] 1. Receiving calls and querying caller information

[1557] The server automatically records calls received by the user and checks the caller information against a database that includes a blacklist of suspicious phone numbers.

[1558] 2. Real-time analysis and automatic response

[1559] The server analyzes the call content in real time and assesses the risk of fraud based on the content of the conversation and the keywords used.

[1560] If the risk of fraud is determined to be high, the server generates an automated response message and sends it to the caller, designed to prevent fraud before it occurs.

[1561] 3. Emotion analysis

[1562] The server uses an emotion engine to analyze the tone, rate and emotional fluctuations of the user's voice during the call to assess the user's emotional state.

[1563] Sentiment analysis results are used to complement fraud risk assessment.

[1564] 4. Notification System

[1565] If a fraudulent call is detected, the notification system will send a real-time notification to the user and their family via SMS, notification app, or email.

[1566] 5. Blacklist Management

[1567] The server stores phone numbers that are deemed suspicious in a blacklist in its database and automatically blocks future calls.

[1568] 6. Continuous Learning System

[1569] The continuous learning system saves newly detected fraud patterns as training data to improve the accuracy of the analysis model.

[1570] Hardware and software used

[1571] 1. Hardware

[1572] Server: Data processing device for analysis and notification

[1573] Smartphone: Records user calls and sends them to a server

[1574] 2. Software

[1575] Call reception app: Records user calls and sends the data to a server

[1576] Generative AI system: Analyzes call content and assesses fraud risk

[1577] Emotion Engine: Analyzes the user's emotional state

[1578] Notification app: Sends notifications about fraud risks

[1579] Specific examples

[1580] 1. Receiving and analyzing calls

[1581] When a user receives a call from an unknown number, the server records the call and checks the caller's number against a database. If the call is not blacklisted, the server analyzes the call in real time.

[1582] 2. Sentiment Analysis and Risk Assessment

[1583] The server uses an emotion engine to analyze the user's emotional state, and the generative AI system analyzes keywords in the call content. The results of the emotion analysis and the call content analysis are combined to assess the fraud risk.

[1584] 3. Auto-replies and notifications

[1585] If the risk of fraud is deemed high, the server generates an automated response message and sends it to the caller, while simultaneously sending real-time notifications to the user and their family members via SMS, notification app, or email.

[1586] Prompt Sentence Examples

[1587] Call content: {Call text}

[1588] Emotional state: {User emotion}

[1589] Criteria: Assess the fraud risk and return a "High Fraud Risk" message if the risk is high.

[1590] In this way, this system reduces the risk of users becoming victims of special fraud and provides an environment in which users can make calls with peace of mind.

[1591] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1592] Step 1:

[1593] When a user receives a call, the device records the call and obtains the caller's phone number. The obtained call data and caller information are sent to the server. The input is the incoming call's voice data and the caller's phone number, and the output is that the data is sent to the server.

[1594] Step 2:

[1595] The server checks the received caller information against the database. If the check finds that the caller is registered on the blacklist, the server automatically blocks the call. The input is the caller's phone number, and the output is the check result and the call blocking process.

[1596] Step 3:

[1597] If the caller is not on the blacklist, the server analyzes the call in real time using a generative AI system that performs keyword analysis and pattern matching on the call content. The input is the audio data of the call, and the output is the fraud risk assessment result.

[1598] Step 4:

[1599] The server uses an emotion engine to analyze the tone, speed, and emotional fluctuations of the user's voice during the call to evaluate the user's emotional state. The input is the voice data of the call, and the output is the evaluation result of the user's emotional state.

[1600] Step 5:

[1601] The server evaluates the risk of fraud by combining the results of keyword analysis by the generative AI system and the results of the emotional state evaluation by the emotion engine. If the risk of fraud is determined to be high, the server generates an automatic response message and sends it to the caller. The input is the analysis results of the generative AI system and the evaluation results of the emotion engine, and the output is the automatic response message.

[1602] Step 6:

[1603] If a fraud risk is detected, the server sends a notification to the user and their family members via SMS, a notification app, or email. The input is the fraud risk assessment result, and the output is the notification message.

[1604] Step 7:

[1605] The server adds any phone numbers it deems suspicious to a blacklist in its database, automatically blocking future calls. The input is the suspicious phone number, and the output is the registration status in the database.

[1606] Step 8:

[1607] The continuous learning system improves accuracy by saving newly detected fraud patterns as training data and updating the analytical models of the generative AI system and emotion engine. The input is the newly detected fraud patterns, and the output is the updated models.

[1608] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1609] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1610] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1611] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1612] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1613] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1614] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1615] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1616] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1617] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1618] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1619] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1620] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1621] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1622] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1623] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1624] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1625] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1626] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1627] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1628] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1629] The following is further disclosed regarding the above embodiment.

[1630] (Claim 1)

[1631] means for receiving the call and querying a database for caller information;

[1632] A means to automatically block calls if the caller is a suspicious number;

[1633] If the caller is not a suspicious number, a method for analyzing the call content in real time and assessing the risk of fraud;

[1634] means for generating and sending an automated response message to a caller if the risk of fraud is determined to be high;

[1635] A means for storing and managing blacklisted telephone numbers in a database;

[1636] A means of notifying users and their families in real time when fraudulent calls are detected;

[1637] A means of learning from newly detected fraud patterns and improving the accuracy of the analytical model; and

[1638] A system including:

[1639] (Claim 2)

[1640] 10. The system of claim 1, further comprising means for playing the generated response message to the caller during the call and disconnecting the call.

[1641] (Claim 3)

[1642] 10. The system of claim 1, including means for notifications to the user and their family members to be sent via SMS, a notification app, or email.

[1643] "Example 1"

[1644] (Claim 1)

[1645] means for receiving the call and querying a database for caller information;

[1646] A means to automatically block calls if the caller is a suspicious number;

[1647] If the caller is not a suspicious number, a method for analyzing the call content in real time and assessing the risk of fraud;

[1648] means for generating and sending an automated response message to a caller if the risk of fraud is determined to be high;

[1649] A means for storing and managing blacklisted telephone numbers in a database;

[1650] A means of notifying users and their families in real time when fraudulent calls are detected;

[1651] A means of learning from newly detected fraud patterns and improving the accuracy of the analytical model; and

[1652] a means for recording and transmitting the call content to the generative AI model system for analysis;

[1653] means for generating an automatic response message based on the analysis result and transmitting the message to the server;

[1654] A system including:

[1655] (Claim 2)

[1656] 10. The system of claim 1, further comprising means for playing the generated response message to the caller during the call and disconnecting the call.

[1657] (Claim 3)

[1658] 10. The system of claim 1, including means for notifications to the user and their family members to be sent via SMS, a notification app, or email.

[1659] "Application Example 1"

[1660] (Claim 1)

[1661] means for receiving the call and querying a database for caller information;

[1662] A means to automatically block calls if the caller is a suspicious number;

[1663] If the caller is not a suspicious number, a method for analyzing the call content in real time and assessing the risk of fraud;

[1664] means for generating and sending an automated response message to a caller if the risk of fraud is determined to be high;

[1665] A means for storing and managing blacklisted telephone numbers in a database;

[1666] A means of notifying users and their families in real time when calls that pose a risk of fraud are detected;

[1667] A means of learning from newly detected fraud patterns and improving the accuracy of the analytical model; and

[1668] A means operated via a mobile communication terminal having a call monitoring function;

[1669] means for transmitting the generated automated response message to a designated recipient, the automated response message having a security function;

[1670] A system including:

[1671] (Claim 2)

[1672] 10. The system of claim 1, further comprising means for playing the generated response message to the caller during the call and disconnecting the call.

[1673] (Claim 3)

[1674] 10. The system of claim 1, including means for sending notifications to the user and their family via SMS, a notification app, or email.

[1675] "Example 2: Combining Emotion Engines"

[1676] (Claim 1)

[1677] means for receiving the call and querying a database for caller information;

[1678] A means to automatically block calls if the caller is a suspicious number;

[1679] If the caller is not a suspicious number, a method for analyzing the call content in real time and assessing the risk of fraud;

[1680] A means of analyzing user sentiment and incorporating it into fraud risk assessments;

[1681] means for generating and sending an automated response message to a caller if the risk of fraud is determined to be high;

[1682] A means for storing and managing blacklisted telephone numbers in a database;

[1683] A means of notifying users and their families in real time when fraudulent calls are detected;

[1684] A means of learning from newly detected fraud patterns and improving the accuracy of the analytical model; and

[1685] A system including:

[1686] (Claim 2)

[1687] 10. The system of claim 1, further comprising means for playing the generated response message to the caller during the call and disconnecting the call.

[1688] (Claim 3)

[1689] 10. The system of claim 1, including means for notifications to the user and their family members to be sent via SMS, a notification app, or email.

[1690] "Application example 2 when combining emotion engines"

[1691] (Claim 1)

[1692] means for receiving the call and querying a database for caller information;

[1693] A means to automatically block calls if the caller is a suspicious number;

[1694] If the caller is not a suspicious number, a method for analyzing the call content in real time and assessing the risk of fraud;

[1695] means for generating and sending an automated response message to a caller if the risk of fraud is determined to be high;

[1696] A means for storing and managing blacklisted telephone numbers in a database;

[1697] A means of notifying users and their families in real time when fraudulent calls are detected;

[1698] A means of learning from newly detected fraud patterns and improving the accuracy of the analytical model; and

[1699] an emotion analysis means for assessing the emotional state of the user;

[1700] a means of complementing fraud risk assessment based on emotional state;

[1701] A system including:

[1702] (Claim 2)

[1703] 10. The system of claim 1, further comprising means for playing the generated response message to the caller during the call and disconnecting the call.

[1704] (Claim 3)

[1705] 10. The system of claim 1, including means for notifications to the user and their family members to be sent via SMS, a notification app, or email. [Explanation of symbols]

[1706] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving the call and querying a database for caller information; A means to automatically block calls if the caller is a suspicious number; If the caller is not a suspicious number, a method for analyzing the call content in real time and assessing the risk of fraud; means for generating and sending an automated response message to a caller if the risk of fraud is determined to be high; A means for storing and managing blacklisted telephone numbers in a database; A means of notifying users and their families in real time when fraudulent calls are detected; A means of learning from newly detected fraud patterns and improving the accuracy of the analytical model; and A system including:

2. 2. The system of claim 1, further comprising means for playing the generated response message to the caller during the call and disconnecting the call.

3. The system of claim 1 , further comprising means for sending notifications to the user and their family via SMS, a notification app, or email.

Images