System

Abstract

An object of the system according to the embodiment is to detect a sign of fraud and issue a warning to a user.SOLUTION: A system includes a mail analysis unit, a call analysis unit, a warning notification unit, and a database update unit. The mail analysis unit analyzes a received mail of a user. The call analysis unit analyzes contents of a telephone call received by a user. The warning notification unit issues a warning to the user when a possibility of fraud is detected. The database update unit records information on the detected fraud in the database and shares the information with other users.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] Conventional technologies do not adequately detect signs of fraud early and warn users, and there is room for improvement.

[0005] The system according to the embodiment aims to detect signs of fraud and issue a warning to the user. [Means for solving the problem]

[0006] The system according to the embodiment includes an email analysis unit, a call analysis unit, a warning notification unit, and a database update unit. The email analysis unit analyzes emails received by the user. The call analysis unit analyzes the content of phone calls received by the user. The warning notification unit issues a warning to the user when a possible fraud is detected. The database update unit records information about the detected fraud in a database and shares it with other users. [Effects of the Invention]

[0007] An embodiment of the system can detect signs of fraud and alert the user. [Brief explanation of the drawings]

[0008] [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. DETAILED DESCRIPTION OF THE INVENTION

[0009] 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.

[0010] First, the terms used in the following description will be explained.

[0011] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] 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.

[0013] 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.

[0014] 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), and Bluetooth (registered trademark).

[0015] 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."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 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.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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).

[0019] 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.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. 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 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. 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.

[0022] 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.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 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.

[0025] 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. The identification processing unit 290 can estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 may have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A fraud detection system according to an embodiment of the present invention uses generative AI to detect signs of fraud and issue a warning to users. This system automatically analyzes fraudulent emails and phone calls and notifies users when there is a possibility of fraud. This allows the fraud detection system to protect users from fraud and prevent them from becoming victims of fraud.

[0029] The fraud detection system according to the embodiment includes an email analysis unit, a call analysis unit, a warning notification unit, and a database update unit. The email analysis unit analyzes emails received by a user. For example, the generation AI analyzes the email content, sender information, linked URLs, and the like to determine whether there is a possibility of fraud. For example, the generation AI analyzes the email content using natural language processing technology to detect signs of fraud. The generation AI can also verify the sender's information and evaluate trustworthiness. Furthermore, the generation AI can analyze linked URLs and evaluate trustworthiness. The call analysis unit analyzes the content of phone calls received by a user. For example, the generation AI converts the call content into text using speech recognition technology and analyzes the text. For example, the generation AI analyzes the call content using natural language processing technology to detect signs of fraud. Furthermore, the generation AI can analyze the audio data of the call content and detect abnormal patterns. Furthermore, the generation AI can analyze the audio data of the call content and estimate the user's emotions using an emotion estimation function. The warning notification unit issues a warning to the user if a possible fraud is detected. For example, if a fraudulent email is detected, a warning message is displayed in the user's email app. Furthermore, if a fraudulent call is detected, a warning sound is sounded during the call or a warning message is displayed after the call ends. The database update unit records the detected fraud information in a database and shares it with other users. For example, if a specific email address or phone number is determined to be related to fraud, that information is added to the database. This allows the fraud detection system according to the embodiment to detect signs of fraud and issue a warning to the user, thereby preventing fraud damage before it occurs. For example, users can become more vigilant against fraudulent emails and phone calls. Furthermore, sharing fraud information with other users can prevent the recurrence of fraud.

[0030] The email analysis unit can analyze the sending time and frequency of emails to detect abnormal patterns. For example, the generation AI in the email analysis unit analyzes the sending time of emails and determines that emails sent outside of normal business hours are abnormal. For example, it identifies emails sent late at night or early in the morning and evaluates the possibility of fraud. The generation AI also analyzes the sending frequency of emails and determines that a large number of emails are sent in a short period of time as abnormal. For example, it evaluates the possibility of fraud when multiple emails are sent from the same sender in a short period of time. The generation AI also analyzes the sending time and frequency of emails in combination to detect abnormal patterns. For example, it identifies emails sent in large numbers during specific time periods and evaluates the possibility of fraud. In this way, by detecting abnormal sending times and frequencies, it is possible to increase the possibility of fraudulent emails.

[0031] The email analysis unit can detect abnormal content or format by comparing it with past email history. In the email analysis unit, for example, the generation AI analyzes past email history to detect expressions or formats that differ from normal email content. For example, it identifies writing styles and formats that differ from normal business emails and evaluates the possibility of fraud. The generation AI also compares it with past email history to detect abnormal links or attachments. For example, it evaluates the possibility of fraud if there are links or attachments that are not included in normal emails. The generation AI also detects abnormal sender information based on past email history. For example, it identifies emails from addresses that are not included in normal sender lists and evaluates the possibility of fraud. In this way, by comparing it with past email history, it is possible to detect abnormal content or format and increase the possibility of fraudulent emails.

[0032] When analyzing the contents of an email, the email analysis unit can also analyze the contents of images and attachments to detect signs of fraud. For example, the generation AI in the email analysis unit analyzes images attached to emails to detect signs of fraud. For example, it evaluates the possibility of fraud if a fake logo or unnatural image editing is used. The generation AI also analyzes the contents of files attached to emails to detect signs of fraud. For example, it evaluates the possibility of fraud if a file contains malware or a suspicious script. The generation AI also analyzes the combined contents of the email body and attachment to detect signs of fraud. For example, it evaluates the possibility of fraud if the contents of the email body and attachment do not match. This makes it easier to detect signs of fraud by analyzing the contents of images and attachments.

[0033] The email analysis unit can reflect feedback from other users in real time and evaluate the likelihood of fraud. In the email analysis unit, for example, the generation AI collects feedback from other users in real time and evaluates the likelihood of fraud. For example, if the same email is reported as fraudulent by multiple users, the likelihood of fraud is increased. The generation AI also analyzes feedback from other users and detects signs of fraud. For example, if a specific sender or link is reported by multiple users, the likelihood of fraud is evaluated. The generation AI also learns fraud patterns based on feedback from other users and evaluates the likelihood of fraud. For example, it detects new signs of fraud based on past feedback data. In this way, the likelihood of fraud can be increased by reflecting feedback from other users in real time.

[0034] The call analysis unit can analyze the start time or duration of a call to detect abnormal patterns. For example, the generation AI in the call analysis unit analyzes the start time of a call and determines that calls made outside of normal business hours are abnormal. For example, calls made late at night or early in the morning are identified and the possibility of fraud is assessed. The generation AI also analyzes the duration of a call and determines that calls that are abnormally long or short are abnormal. For example, calls that differ significantly from normal call lengths are identified and the possibility of fraud is assessed. The generation AI also analyzes the combination of the start time and duration of a call to detect abnormal patterns. For example, calls that are concentrated during certain hours are identified and the possibility of fraud is assessed. In this way, by analyzing the start time and duration of a call, abnormal patterns can be detected and the possibility of fraudulent calls can be increased.

[0035] The call analysis unit can detect abnormal content or formats by comparing with past call history. In the call analysis unit, for example, the generation AI analyzes past call history to detect expressions or formats that differ from normal call content. For example, it identifies writing styles and formats that are not found in normal calls and evaluates the possibility of fraud. The generation AI also compares with past call history to detect abnormal content or formats. For example, it evaluates the possibility of fraud if there are requests or instructions that are not included in normal calls. Furthermore, a system is built in which the generation AI detects abnormal call content or formats based on past call history. For example, it evaluates the possibility of fraud if the content matches the content of calls in which fraudulent acts have been reported in the past. In this way, by comparing with past call history, abnormal content or formats can be detected and the possibility of a fraudulent call can be increased.

[0036] The call analysis unit can analyze background sounds or noises during a call and detect signs of fraud. In the call analysis unit, for example, the generation AI analyzes background sounds during a call and evaluates the possibility of fraud if abnormal sounds are included. For example, the possibility of fraud is evaluated if there is background sounds that are not audible in normal calls. The generation AI also analyzes noise during a call and evaluates the possibility of fraud if abnormal noise is included. For example, the possibility of fraud is evaluated if there is noise that is not audible in normal calls. In addition, a system is constructed in which the generation AI analyzes a combination of background sounds and noise during a call and evaluates the possibility of fraud if abnormal sounds are included. For example, the possibility of fraud is evaluated if there is abnormal sounds that are not audible in normal calls. This makes it easier to detect signs of fraud by analyzing background sounds and noise during a call.

[0037] The call analysis unit can reflect feedback from other users in real time and evaluate the possibility of fraud. In the call analysis unit, for example, the generation AI collects feedback from other users in real time and evaluates the possibility of fraud. For example, if the same call content is reported as fraud by multiple users, the possibility of fraud is increased. The generation AI also analyzes feedback from other users and detects signs of fraud. For example, if specific call content is reported by multiple users, the possibility of fraud is evaluated. The generation AI also learns fraud patterns based on feedback from other users and evaluates the possibility of fraud. For example, it detects new signs of fraud based on past feedback data. In this way, the possibility of fraud can be increased by reflecting feedback from other users in real time.

[0038] The call analysis unit can also analyze the tone and speed of the voice when converting the content of a call into text using speech recognition technology to detect abnormal patterns. For example, when the generation AI converts the content of a call into text using speech recognition technology, the call analysis unit analyzes the tone of the voice and evaluates the possibility of fraud if an abnormal tone is included. For example, if there is a strong tone that is not heard in normal calls, the possibility of fraud is evaluated. In addition, when the generation AI converts the content of a call into text using speech recognition technology, the call analysis unit analyzes the speed of the voice and evaluates the possibility of fraud if an abnormal speed is included. For example, if there is a fast or slow speed that is not seen in normal calls, the possibility of fraud is evaluated. In addition, when the generation AI converts the content of a call into text using speech recognition technology, the system analyzes the tone and speed of the voice in combination to detect abnormal patterns. For example, if there is an abnormal tone and speed that are not seen in normal calls, the possibility of fraud is evaluated. In this way, by analyzing the tone and speed of the voice, abnormal patterns can be detected and the possibility of a fraudulent call can be increased.

[0039] The call analysis unit can detect abnormal behavioral patterns by referencing the call recipient's past behavioral history. For example, the call analysis unit uses a generation AI to analyze the call recipient's past behavioral history and detect call content that differs from normal behavioral patterns. For example, if there are requests or instructions that are not seen in normal calls, the possibility of fraud is assessed. The generation AI also compares this with the call recipient's past behavioral history to detect abnormal behavioral patterns. For example, it identifies calls from call recipients with behavioral patterns that are not seen in normal calls and assesses the possibility of fraud. The generation AI also builds a system that detects abnormal behavioral patterns based on the call recipient's past behavioral history. For example, it identifies calls from call recipients who have been reported to have committed fraud in the past and assesses the possibility of fraud. In this way, by referencing the call recipient's past behavioral history, it is possible to detect abnormal behavioral patterns and increase the possibility of fraudulent calls.

[0040] When analyzing the content of a call, the call analysis unit can simultaneously analyze the content of other communication means to detect signs of fraud. For example, when the generation AI analyzes the content of a call, the call analysis unit simultaneously analyzes the content of other communication means (e.g., SMS or chat) to build a system that detects signs of fraud. For example, if the same fraudulent message is sent via multiple communication means, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it references the content of other communication means to detect signs of fraud. For example, if the content of the call matches the content of an SMS, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it combines and analyzes the content of other communication means to build a system that detects signs of fraud. For example, if the content of the call matches the content of a chat, the possibility of fraud is assessed. In this way, by simultaneously analyzing the content of other communication means, it is possible to more easily detect signs of fraud.

[0041] When analyzing call content, the call analysis unit can detect abnormal content by referring to the user's past behavioral history and preferences. In the call analysis unit, for example, the generation AI analyzes the user's past behavioral history and detects call content that differs from normal behavioral patterns. For example, it identifies content and formats that are not seen in normal calls and evaluates the possibility of fraud. The generation AI also analyzes the user's preferences and detects call content that differs from normal preferences. For example, it evaluates the possibility of fraud when there are requests or instructions that are not included in normal calls. The generation AI also builds a system that combines and analyzes the user's past behavioral history and preferences to detect abnormal call content. For example, it identifies content and formats that differ from normal behavioral patterns and evaluates the possibility of fraud. In this way, by referring to the user's past behavioral history and preferences, it is possible to detect abnormal content and increase the possibility of fraudulent calls.

[0042] The database update unit records information about detected frauds in a database and can share it with other users. For example, the database update unit records information about frauds detected by the generation AI in a database and shares it with other users. For example, if a specific email address or phone number is determined to be related to fraud, that information is added to the database. This allows other users to be warned if they encounter the same fraud. In this way, by recording information about detected frauds in a database and sharing it with other users, it is possible to prevent frauds from occurring again.

[0043] If a specific email address or phone number is determined to be related to fraud, the database update unit can add that information to the database. For example, if the generation AI determines that a specific email address or phone number is related to fraud, the database update unit adds that information to the database. For example, email addresses and phone numbers that have been reported as fraudulent in the past can be recorded in the database and shared with other users. This allows other users to be warned if they encounter the same scam. As a result, if a specific email address or phone number is determined to be related to fraud, adding that information to the database allows other users to be warned if they encounter the same scam.

[0044] The database update unit can show typical examples of fraudulent emails and phone calls and explain countermeasures against them. For example, the database update unit uses a generation AI to collect typical examples of fraudulent emails and phone calls and explain countermeasures against them. For example, the database update unit analyzes typical fraud patterns based on a database of past cases and provides countermeasures to the user. This allows the user to deepen their knowledge of fraud and develop the ability to spot fraud themselves. This allows the user to deepen their knowledge of fraud and develop the ability to spot fraud themselves by showing typical examples of fraudulent emails and phone calls and explaining countermeasures against them.

[0045] The warning notification unit can display a warning message in the user's email app when a fraudulent email is detected. For example, when the generation AI detects a fraudulent email, the warning notification unit displays a warning message in the user's email app. For example, when an email that is highly likely to be fraudulent is received, a warning message is displayed to the user to warn them. This allows the user to be more vigilant against fraudulent emails. By displaying a warning message in the user's email app when a fraudulent email is detected, the user can be more vigilant against fraudulent emails.

[0046] The warning notification unit can sound a warning sound during a call or display a warning message after the call ends if a fraudulent call is detected. For example, if the generation AI detects a fraudulent call, the warning notification unit can sound a warning sound during a call or display a warning message after the call ends. For example, if a call that is likely to be fraudulent is being made, a warning sound can be sounded to alert the user. Also, a warning message can be displayed after the call ends to notify the user of the possibility of fraud. This allows the user to be more vigilant against fraudulent calls. This allows the user to be more vigilant against fraudulent calls by sounding a warning sound during a call or displaying a warning message after the call ends if a fraudulent call is detected.

[0047] When analyzing email sender information, the email analysis unit can detect abnormal behavioral patterns by referring to the sender's past behavioral history. For example, the generation AI in the email analysis unit analyzes email sender information and detects abnormal behavioral patterns by referring to the past behavioral history. For example, it identifies emails from addresses that are not included in the normal sender list and evaluates the possibility of fraud. The generation AI also analyzes email sender information and compares it with the past behavioral history to detect abnormal behavioral patterns. For example, it identifies emails from senders with behavioral patterns that differ from normal senders and evaluates the possibility of fraud. The generation AI also builds a system that analyzes email sender information and detects abnormal behavioral patterns based on the past behavioral history. For example, it identifies emails from senders who have been reported to have committed fraud in the past and evaluates the possibility of fraud. As a result, when analyzing email sender information, it can detect abnormal behavioral patterns by referring to the sender's past behavioral history, thereby increasing the possibility of fraudulent emails.

[0048] When analyzing the URL linked from an email, the email analysis unit can evaluate its trustworthiness by referring to the URL's domain information and registration information. For example, the generation AI analyzes the URL linked from an email and evaluates its trustworthiness by referring to the domain information. For example, it identifies links from domains with low reliability and evaluates the possibility of fraud. The generation AI also analyzes the URL linked from an email and evaluates its trustworthiness by referring to the registration information. For example, it identifies URLs with unclear registration information and evaluates the possibility of fraud. The generation AI also analyzes the URL linked from an email and builds a system that evaluates its trustworthiness by combining the domain information and registration information. For example, it identifies domains with low reliability and URLs with unclear registration information and evaluates the possibility of fraud. As a result, when analyzing the URL linked from an email, it can evaluate its trustworthiness by referring to the URL's domain information and registration information, thereby increasing the possibility of it being a fraudulent email.

[0049] When analyzing the content of an email, the email analysis unit can simultaneously analyze the content of other communication means to detect signs of fraud. For example, when the generation AI analyzes the content of an email, the email analysis unit simultaneously analyzes the content of other communication means (e.g., SMS and chat) to build a system that detects signs of fraud. For example, if a fraudulent message with the same content is sent via multiple communication means, the possibility of fraud is evaluated. In addition, when the generation AI analyzes the content of an email, it references the content of other communication means to detect signs of fraud. For example, if the content of an email and an SMS match, the possibility of fraud is evaluated. In addition, when the generation AI analyzes the content of an email, it combines and analyzes the content of other communication means to build a system that detects signs of fraud. For example, if the content of an email and a chat match, the possibility of fraud is evaluated. In this way, by simultaneously analyzing the content of other communication means, it is possible to more easily detect signs of fraud.

[0050] When analyzing the content of an email, the email analysis unit can refer to the user's past behavioral history and preferences to detect abnormal content. For example, the email analysis unit uses a generation AI to analyze the user's past behavioral history and detect email content that differs from normal behavioral patterns. For example, it identifies content and formats not found in normal emails and evaluates the possibility of fraud. The generation AI also analyzes the user's preferences and detects email content that differs from normal preferences. For example, it evaluates the possibility of fraud if the email contains links or attachments not found in normal emails. The generation AI also builds a system that combines and analyzes the user's past behavioral history and preferences to detect abnormal email content. For example, it identifies content and formats that differ from normal behavioral patterns and evaluates the possibility of fraud. In this way, by referring to the user's past behavioral history and preferences, it is possible to detect abnormal content and increase the possibility of fraudulent emails.

[0051] When analyzing the content of a call, the call analysis unit can simultaneously analyze the content of other communication means to detect signs of fraud. For example, when the generation AI analyzes the content of a call, the call analysis unit simultaneously analyzes the content of other communication means (e.g., SMS or chat) to build a system that detects signs of fraud. For example, if the same fraudulent message is sent via multiple communication means, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it references the content of other communication means to detect signs of fraud. For example, if the content of the call matches the content of an SMS, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it combines and analyzes the content of other communication means to build a system that detects signs of fraud. For example, if the content of the call matches the content of a chat, the possibility of fraud is assessed. In this way, by simultaneously analyzing the content of other communication means, it is possible to more easily detect signs of fraud.

[0052] When analyzing call content, the call analysis unit can detect abnormal content by referring to the user's past behavioral history and preferences. In the call analysis unit, for example, the generation AI analyzes the user's past behavioral history and detects call content that differs from normal behavioral patterns. For example, it identifies content and formats that are not seen in normal calls and evaluates the possibility of fraud. The generation AI also analyzes the user's preferences and detects call content that differs from normal preferences. For example, it evaluates the possibility of fraud when there are requests or instructions that are not included in normal calls. The generation AI also builds a system that combines and analyzes the user's past behavioral history and preferences to detect abnormal call content. For example, it identifies content and formats that differ from normal behavioral patterns and evaluates the possibility of fraud. In this way, by referring to the user's past behavioral history and preferences, it is possible to detect abnormal content and increase the possibility of fraudulent calls.

[0053] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0054] The fraud detection system may further include a behavior analysis unit that analyzes a user's behavioral history. The behavior analysis unit, for example, analyzes a user's past online behavior and purchase history to detect behavior that differs from normal behavioral patterns. For example, the possibility of fraud is assessed when a high-value purchase that differs from normal purchasing patterns is made. The behavior analysis unit also analyzes a user's past browsing history to detect access to sites that differ from normal browsing patterns. For example, the possibility of fraud is assessed when a suspicious site that differs from normal browsing sites is accessed. Furthermore, the behavior analysis unit analyzes a user's past login history to detect abnormal login patterns. For example, the possibility of fraud is assessed when a login occurs at a time or location that differs from normal login times. In this way, analyzing a user's behavioral history can make it easier to detect signs of fraud.

[0055] The fraud detection system may further include a biometric authentication unit that analyzes the user's biometric information. The biometric authentication unit, for example, analyzes the user's fingerprint or face authentication data and evaluates the possibility of fraud if the data differs from a normal authentication pattern. For example, if a fingerprint different from a normal fingerprint pattern is detected, the possibility of fraud is evaluated. The biometric authentication unit also analyzes the user's voiceprint data and evaluates the possibility of fraud if the data differs from a normal voiceprint pattern. For example, if a voiceprint different from a normal voiceprint is detected, the possibility of fraud is evaluated. The biometric authentication unit also analyzes the user's iris data and evaluates the possibility of fraud if the data differs from a normal iris pattern. For example, if an iris different from a normal iris pattern is detected, the possibility of fraud is evaluated. This makes it easier to detect signs of fraud by analyzing the biometric information.

[0056] The fraud detection system may further include a location information analysis unit that analyzes the user's location information. The location information analysis unit, for example, analyzes the user's current location and evaluates the possibility of fraud if the user is in a location different from their usual range of activity. For example, it evaluates the possibility of fraud if the user is in a country or region different from their usual range of activity. The location information analysis unit also analyzes the user's past movement history and detects abnormal movement patterns. For example, it evaluates the possibility of fraud if there is a sudden movement that differs from the usual movement pattern. Furthermore, the location information analysis unit combines the user's location information with other data and analyzes it to detect abnormal patterns. For example, it evaluates the possibility of fraud if there is a login from a location different from the usual range of activity. In this way, analyzing the location information can make it easier to detect signs of fraud.

[0057] The fraud detection system may further include a device analysis unit that analyzes the user's device information. The device analysis unit, for example, analyzes information about the device used by the user and evaluates the possibility of fraud if it is different from the usual device. For example, it evaluates the possibility of fraud if there is access from a device different from the device normally used. The device analysis unit also analyzes the IP address of the user's device and evaluates the possibility of fraud if it is different from the usual IP address. For example, it evaluates the possibility of fraud if there is access from an IP address different from the usual IP address. The device analysis unit also analyzes the MAC address of the user's device and evaluates the possibility of fraud if it is different from the usual MAC address. For example, it evaluates the possibility of fraud if there is access from a MAC address different from the usual MAC address. In this way, analyzing the device information makes it easier to detect signs of fraud.

[0058] The fraud detection system may further include a social media analysis unit that analyzes the user's social media activity. The social media analysis unit, for example, analyzes the content of the user's posts on social media and evaluates the possibility of fraud if the content differs from the user's normal posting pattern. For example, the social media analysis unit evaluates the possibility of fraud if a suspicious post different from the normal posting content is made. The social media analysis unit also analyzes the user's social media friend list and evaluates the possibility of fraud if the content differs from the normal friend list. For example, the social media analysis unit evaluates the possibility of fraud if a suspicious account not included in the normal friend list is added. The social media analysis unit also analyzes the content of the user's messages on social media and evaluates the possibility of fraud if the content differs from the normal message content. For example, the social media analysis unit evaluates the possibility of fraud if a suspicious message different from the normal message content is sent. In this way, analyzing social media activity can make it easier to detect signs of fraud.

[0059] The processing flow of the first embodiment will be briefly explained below.

[0060] Step 1: The email analysis unit analyzes the email received by the user. The generation AI analyzes the email content, sender information, linked URLs, etc. to determine whether there is a possibility of fraud. For example, it can use natural language processing technology to analyze the email content and detect signs of fraud. It can also verify the sender's information and evaluate their trustworthiness. It can also analyze linked URLs and evaluate their trustworthiness. Step 2: The call analysis unit analyzes the content of the call received by the user. The generation AI converts the content of the call into text using voice recognition technology and analyzes the text. For example, it can analyze the content of the call using natural language processing technology to detect signs of fraud. It can also analyze the audio data of the call and detect abnormal patterns. It can also estimate the user's emotions using an emotion estimation function. Step 3: The warning notification unit issues a warning to the user if a possible fraud is detected. For example, if a fraudulent email is detected, a warning message is displayed in the user's email app. Also, if a fraudulent phone call is detected, a warning sound is played during the call or a warning message is displayed after the call ends. Step 4: The database updater records the detected fraud information in a database and shares it with other users. For example, if a specific email address or phone number is determined to be related to fraud, that information is added to the database.

[0061] (Example 2) A fraud detection system according to an embodiment of the present invention uses generative AI to detect signs of fraud and issue a warning to users. This system automatically analyzes fraudulent emails and phone calls and notifies users when there is a possibility of fraud. This allows the fraud detection system to protect users from fraud and prevent them from becoming victims of fraud.

[0062] The fraud detection system according to the embodiment includes an email analysis unit, a call analysis unit, a warning notification unit, and a database update unit. The email analysis unit analyzes emails received by a user. For example, the generation AI analyzes the email content, sender information, linked URLs, and the like to determine whether there is a possibility of fraud. For example, the generation AI analyzes the email content using natural language processing technology to detect signs of fraud. The generation AI can also verify the sender's information and evaluate trustworthiness. Furthermore, the generation AI can analyze linked URLs and evaluate trustworthiness. The call analysis unit analyzes the content of phone calls received by a user. For example, the generation AI converts the call content into text using speech recognition technology and analyzes the text. For example, the generation AI analyzes the call content using natural language processing technology to detect signs of fraud. Furthermore, the generation AI can analyze the audio data of the call content and detect abnormal patterns. Furthermore, the generation AI can analyze the audio data of the call content and estimate the user's emotions using an emotion estimation function. The warning notification unit issues a warning to the user if a possible fraud is detected. For example, if a fraudulent email is detected, a warning message is displayed in the user's email app. Furthermore, if a fraudulent call is detected, a warning sound is sounded during the call or a warning message is displayed after the call ends. The database update unit records the detected fraud information in a database and shares it with other users. For example, if a specific email address or phone number is determined to be related to fraud, that information is added to the database. This allows the fraud detection system according to the embodiment to detect signs of fraud and issue a warning to the user, thereby preventing fraud damage before it occurs. For example, users can become more vigilant against fraudulent emails and phone calls. Furthermore, sharing fraud information with other users can prevent the recurrence of fraud.

[0063] The email analysis unit can analyze the sending time and frequency of emails to detect abnormal patterns. For example, the generation AI in the email analysis unit analyzes the sending time of emails and determines that emails sent outside of normal business hours are abnormal. For example, it identifies emails sent late at night or early in the morning and evaluates the possibility of fraud. The generation AI also analyzes the sending frequency of emails and determines that a large number of emails are sent in a short period of time as abnormal. For example, it evaluates the possibility of fraud when multiple emails are sent from the same sender in a short period of time. The generation AI also analyzes the sending time and frequency of emails in combination to detect abnormal patterns. For example, it identifies emails sent in large numbers during specific time periods and evaluates the possibility of fraud. In this way, by detecting abnormal sending times and frequencies, it is possible to increase the possibility of fraudulent emails.

[0064] The email analysis unit can detect abnormal content or format by comparing it with past email history. In the email analysis unit, for example, the generation AI analyzes past email history to detect expressions or formats that differ from normal email content. For example, it identifies writing styles and formats that differ from normal business emails and evaluates the possibility of fraud. The generation AI also compares it with past email history to detect abnormal links or attachments. For example, it evaluates the possibility of fraud if there are links or attachments that are not included in normal emails. The generation AI also detects abnormal sender information based on past email history. For example, it identifies emails from addresses that are not included in normal sender lists and evaluates the possibility of fraud. In this way, by comparing it with past email history, it is possible to detect abnormal content or format and increase the possibility of fraudulent emails.

[0065] The email analysis unit uses an emotion estimation function to estimate the user's emotions from the content of the email, and can increase the likelihood of fraud if there are abnormal emotional fluctuations. In the email analysis unit, for example, the generation AI analyzes the content of the email and uses the emotion estimation function to estimate the user's emotions. For example, if the content of the email causes anxiety or fear, it evaluates the likelihood of fraud. The generation AI also analyzes the content of the email and uses the emotion estimation function to detect abnormal emotional fluctuations. For example, it evaluates the likelihood of fraud if the email contains strong emotional expressions not seen in normal emails. The generation AI also analyzes the content of the email and uses the emotion estimation function to monitor the user's emotional fluctuations in real time. For example, it evaluates the likelihood of fraud if the user's emotions change suddenly when they open the email. In this way, the emotion estimation function can detect abnormal emotional fluctuations and increase the likelihood of a fraudulent email.

[0066] When analyzing the contents of an email, the email analysis unit can also analyze the contents of images and attachments to detect signs of fraud. For example, the generation AI in the email analysis unit analyzes images attached to emails to detect signs of fraud. For example, it evaluates the possibility of fraud if a fake logo or unnatural image editing is used. The generation AI also analyzes the contents of files attached to emails to detect signs of fraud. For example, it evaluates the possibility of fraud if a file contains malware or a suspicious script. The generation AI also analyzes the combined contents of the email body and attachment to detect signs of fraud. For example, it evaluates the possibility of fraud if the contents of the email body and attachment do not match. This makes it easier to detect signs of fraud by analyzing the contents of images and attachments.

[0067] The email analysis unit can reflect feedback from other users in real time and evaluate the likelihood of fraud. In the email analysis unit, for example, the generation AI collects feedback from other users in real time and evaluates the likelihood of fraud. For example, if the same email is reported as fraudulent by multiple users, the likelihood of fraud is increased. The generation AI also analyzes feedback from other users and detects signs of fraud. For example, if a specific sender or link is reported by multiple users, the likelihood of fraud is evaluated. The generation AI also learns fraud patterns based on feedback from other users and evaluates the likelihood of fraud. For example, it detects new signs of fraud based on past feedback data. In this way, the likelihood of fraud can be increased by reflecting feedback from other users in real time.

[0068] The email analysis unit uses an emotion estimation function to monitor the user's emotions in real time when they open an email and can issue a warning if there is an abnormal emotional reaction. For example, the email analysis unit uses a generation AI to monitor the user's emotions in real time when they open an email and issue a warning if there is an abnormal emotional reaction. For example, a warning is displayed if the user feels anxiety or fear. The generation AI also analyzes the user's emotional reaction and issues a warning if there is an abnormal emotional fluctuation. For example, a warning is displayed if there is a strong emotional reaction not seen in normal emails. A system can also be built in which the generation AI monitors the user's emotional reaction in real time and issues a warning if there is an abnormal emotional fluctuation. For example, it analyzes the user's facial expressions and voice and issues a warning if there is an abnormal emotional reaction. This allows the system to monitor the user's emotions in real time when they open an email and issue a warning if there is an abnormal emotional reaction, thereby increasing the possibility of fraud.

[0069] The call analysis unit can analyze the start time or duration of a call to detect abnormal patterns. For example, the generation AI in the call analysis unit analyzes the start time of a call and determines that calls made outside of normal business hours are abnormal. For example, calls made late at night or early in the morning are identified and the possibility of fraud is assessed. The generation AI also analyzes the duration of a call and determines that calls that are abnormally long or short are abnormal. For example, calls that differ significantly from normal call lengths are identified and the possibility of fraud is assessed. The generation AI also analyzes the combination of the start time and duration of a call to detect abnormal patterns. For example, calls that are concentrated during certain hours are identified and the possibility of fraud is assessed. In this way, by analyzing the start time and duration of a call, abnormal patterns can be detected and the possibility of fraudulent calls can be increased.

[0070] The call analysis unit can detect abnormal content or formats by comparing with past call history. In the call analysis unit, for example, the generation AI analyzes past call history to detect expressions or formats that differ from normal call content. For example, it identifies writing styles and formats that are not found in normal calls and evaluates the possibility of fraud. The generation AI also compares with past call history to detect abnormal content or formats. For example, it evaluates the possibility of fraud if there are requests or instructions that are not included in normal calls. Furthermore, a system is built in which the generation AI detects abnormal call content or formats based on past call history. For example, it evaluates the possibility of fraud if the content matches the content of calls in which fraudulent acts have been reported in the past. In this way, by comparing with past call history, abnormal content or formats can be detected and the possibility of a fraudulent call can be increased.

[0071] The call analysis unit uses an emotion estimation function to estimate the user's emotions from the call content, and can increase the possibility of fraud if there are abnormal emotional fluctuations. In the call analysis unit, for example, the generation AI analyzes the call content and uses the emotion estimation function to estimate the user's emotions. For example, if the call content causes anxiety or fear, the possibility of fraud is evaluated. The generation AI also analyzes the call content and uses the emotion estimation function to detect abnormal emotional fluctuations. For example, if the call content contains strong emotional expressions not seen in normal calls, the possibility of fraud is evaluated. The generation AI also analyzes the call content and uses the emotion estimation function to monitor the user's emotional fluctuations in real time. For example, if the user's emotions change suddenly during the call, the possibility of fraud is evaluated. In this way, the emotion estimation function can detect abnormal emotional fluctuations and increase the possibility of a fraudulent call.

[0072] The call analysis unit can analyze background sounds or noises during a call and detect signs of fraud. In the call analysis unit, for example, the generation AI analyzes background sounds during a call and evaluates the possibility of fraud if abnormal sounds are included. For example, the possibility of fraud is evaluated if there is background sounds that are not audible in normal calls. The generation AI also analyzes noise during a call and evaluates the possibility of fraud if abnormal noise is included. For example, the possibility of fraud is evaluated if there is noise that is not audible in normal calls. In addition, a system is constructed in which the generation AI analyzes a combination of background sounds and noise during a call and evaluates the possibility of fraud if abnormal sounds are included. For example, the possibility of fraud is evaluated if there is abnormal sounds that are not audible in normal calls. This makes it easier to detect signs of fraud by analyzing background sounds and noise during a call.

[0073] The call analysis unit can reflect feedback from other users in real time and evaluate the possibility of fraud. In the call analysis unit, for example, the generation AI collects feedback from other users in real time and evaluates the possibility of fraud. For example, if the same call content is reported as fraud by multiple users, the possibility of fraud is increased. The generation AI also analyzes feedback from other users and detects signs of fraud. For example, if specific call content is reported by multiple users, the possibility of fraud is evaluated. The generation AI also learns fraud patterns based on feedback from other users and evaluates the possibility of fraud. For example, it detects new signs of fraud based on past feedback data. In this way, the possibility of fraud can be increased by reflecting feedback from other users in real time.

[0074] The call analysis unit uses an emotion estimation function to monitor the emotions felt by the user during a call in real time and issue a warning if an abnormal emotional reaction is detected. For example, the call analysis unit uses a generation AI to monitor the emotions felt by the user during a call in real time and issue a warning if an abnormal emotional reaction is detected. For example, a warning is displayed if the user feels anxiety or fear. The generation AI also analyzes the user's emotional reaction and issues a warning if there are abnormal emotional fluctuations. For example, a warning is displayed if there is a strong emotional reaction not seen in a normal call. Furthermore, a system is constructed in which the generation AI monitors the user's emotional reaction in real time and issues a warning if there are abnormal emotional fluctuations. For example, it analyzes the user's facial expressions and voice and issues a warning if there is an abnormal emotional reaction. This allows the system to monitor the emotions felt by the user during a call in real time and issue a warning if there is an abnormal emotional reaction, thereby increasing the possibility of fraud.

[0075] The call analysis unit can also analyze the tone and speed of the voice when converting the content of a call into text using speech recognition technology to detect abnormal patterns. For example, when the generation AI converts the content of a call into text using speech recognition technology, the call analysis unit analyzes the tone of the voice and evaluates the possibility of fraud if an abnormal tone is included. For example, if there is a strong tone that is not heard in normal calls, the possibility of fraud is evaluated. In addition, when the generation AI converts the content of a call into text using speech recognition technology, the call analysis unit analyzes the speed of the voice and evaluates the possibility of fraud if an abnormal speed is included. For example, if there is a fast or slow speed that is not seen in normal calls, the possibility of fraud is evaluated. In addition, when the generation AI converts the content of a call into text using speech recognition technology, the system analyzes the tone and speed of the voice in combination to detect abnormal patterns. For example, if there is an abnormal tone and speed that are not seen in normal calls, the possibility of fraud is evaluated. In this way, by analyzing the tone and speed of the voice, abnormal patterns can be detected and the possibility of a fraudulent call can be increased.

[0076] The call analysis unit can detect abnormal behavioral patterns by referencing the call recipient's past behavioral history. For example, the call analysis unit uses a generation AI to analyze the call recipient's past behavioral history and detect call content that differs from normal behavioral patterns. For example, if there are requests or instructions that are not seen in normal calls, the possibility of fraud is assessed. The generation AI also compares this with the call recipient's past behavioral history to detect abnormal behavioral patterns. For example, it identifies calls from call recipients with behavioral patterns that are not seen in normal calls and assesses the possibility of fraud. The generation AI also builds a system that detects abnormal behavioral patterns based on the call recipient's past behavioral history. For example, it identifies calls from call recipients who have been reported to have committed fraud in the past and assesses the possibility of fraud. In this way, by referencing the call recipient's past behavioral history, it is possible to detect abnormal behavioral patterns and increase the possibility of fraudulent calls.

[0077] The call analysis unit uses an emotion estimation function to estimate the emotions of the call recipient and can increase the likelihood of fraud if there are abnormal emotional fluctuations. In the call analysis unit, for example, the generation AI estimates the emotions of the call recipient and increases the likelihood of fraud if there are abnormal emotional fluctuations. For example, the call is evaluated as likely to be fraud if it contains strong emotional expressions not seen in normal calls. In addition, a system is constructed in which the generation AI estimates the emotions of the call recipient and detects abnormal emotional fluctuations. For example, the possibility of fraud is evaluated if the call recipient is feeling anxious or fearful. In addition, a system is constructed in which the generation AI estimates the emotions of the call recipient and issues a warning if there are abnormal emotional fluctuations. For example, the possibility of fraud is evaluated if the call recipient's emotions change suddenly. In this way, by estimating the emotions of the call recipient, abnormal emotional fluctuations can be detected and the possibility of a fraudulent call can be increased.

[0078] When analyzing the content of a call, the call analysis unit can simultaneously analyze the content of other communication means to detect signs of fraud. For example, when the generation AI analyzes the content of a call, the call analysis unit simultaneously analyzes the content of other communication means (e.g., SMS or chat) to build a system that detects signs of fraud. For example, if the same fraudulent message is sent via multiple communication means, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it references the content of other communication means to detect signs of fraud. For example, if the content of the call matches the content of an SMS, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it combines and analyzes the content of other communication means to build a system that detects signs of fraud. For example, if the content of the call matches the content of a chat, the possibility of fraud is assessed. In this way, by simultaneously analyzing the content of other communication means, it is possible to more easily detect signs of fraud.

[0079] When analyzing call content, the call analysis unit can detect abnormal content by referring to the user's past behavioral history and preferences. In the call analysis unit, for example, the generation AI analyzes the user's past behavioral history and detects call content that differs from normal behavioral patterns. For example, it identifies content and formats that are not seen in normal calls and evaluates the possibility of fraud. The generation AI also analyzes the user's preferences and detects call content that differs from normal preferences. For example, it evaluates the possibility of fraud when there are requests or instructions that are not included in normal calls. The generation AI also builds a system that combines and analyzes the user's past behavioral history and preferences to detect abnormal call content. For example, it identifies content and formats that differ from normal behavioral patterns and evaluates the possibility of fraud. In this way, by referring to the user's past behavioral history and preferences, it is possible to detect abnormal content and increase the possibility of fraudulent calls.

[0080] The call analysis unit uses an emotion estimation function to monitor the emotions felt by the user during a call in real time and issue a warning if an abnormal emotional reaction is detected. For example, the call analysis unit uses a generation AI to monitor the emotions felt by the user during a call in real time and issue a warning if an abnormal emotional reaction is detected. For example, a warning is displayed if the user feels anxiety or fear. The generation AI also analyzes the user's emotional reaction and issues a warning if there are abnormal emotional fluctuations. For example, a warning is displayed if there is a strong emotional reaction not seen in a normal call. Furthermore, a system is constructed in which the generation AI monitors the user's emotional reaction in real time and issues a warning if there are abnormal emotional fluctuations. For example, it analyzes the user's facial expressions and voice and issues a warning if there is an abnormal emotional reaction. This allows the system to monitor the emotions felt by the user during a call in real time and issue a warning if there is an abnormal emotional reaction, thereby increasing the possibility of fraud.

[0081] The database update unit records information about detected frauds in a database and can share it with other users. For example, the database update unit records information about frauds detected by the generation AI in a database and shares it with other users. For example, if a specific email address or phone number is determined to be related to fraud, that information is added to the database. This allows other users to be warned if they encounter the same fraud. In this way, by recording information about detected frauds in a database and sharing it with other users, it is possible to prevent frauds from occurring again.

[0082] If a specific email address or phone number is determined to be related to fraud, the database update unit can add that information to the database. For example, if the generation AI determines that a specific email address or phone number is related to fraud, the database update unit adds that information to the database. For example, email addresses and phone numbers that have been reported as fraudulent in the past can be recorded in the database and shared with other users. This allows other users to be warned if they encounter the same scam. As a result, if a specific email address or phone number is determined to be related to fraud, adding that information to the database allows other users to be warned if they encounter the same scam.

[0083] The database update unit can show typical examples of fraudulent emails and phone calls and explain countermeasures against them. For example, the database update unit uses a generation AI to collect typical examples of fraudulent emails and phone calls and explain countermeasures against them. For example, the database update unit analyzes typical fraud patterns based on a database of past cases and provides countermeasures to the user. This allows the user to deepen their knowledge of fraud and develop the ability to spot fraud themselves. This allows the user to deepen their knowledge of fraud and develop the ability to spot fraud themselves by showing typical examples of fraudulent emails and phone calls and explaining countermeasures against them.

[0084] The warning notification unit can display a warning message in the user's email app when a fraudulent email is detected. For example, when the generation AI detects a fraudulent email, the warning notification unit displays a warning message in the user's email app. For example, when an email that is highly likely to be fraudulent is received, a warning message is displayed to the user to warn them. This allows the user to be more vigilant against fraudulent emails. By displaying a warning message in the user's email app when a fraudulent email is detected, the user can be more vigilant against fraudulent emails.

[0085] The warning notification unit can sound a warning sound during a call or display a warning message after the call ends if a fraudulent call is detected. For example, if the generation AI detects a fraudulent call, the warning notification unit can sound a warning sound during a call or display a warning message after the call ends. For example, if a call that is likely to be fraudulent is being made, a warning sound can be sounded to alert the user. Also, a warning message can be displayed after the call ends to notify the user of the possibility of fraud. This allows the user to be more vigilant against fraudulent calls. This allows the user to be more vigilant against fraudulent calls by sounding a warning sound during a call or displaying a warning message after the call ends if a fraudulent call is detected.

[0086] When analyzing email sender information, the email analysis unit can detect abnormal behavioral patterns by referring to the sender's past behavioral history. For example, the generation AI in the email analysis unit analyzes email sender information and detects abnormal behavioral patterns by referring to the past behavioral history. For example, it identifies emails from addresses that are not included in the normal sender list and evaluates the possibility of fraud. The generation AI also analyzes email sender information and compares it with the past behavioral history to detect abnormal behavioral patterns. For example, it identifies emails from senders with behavioral patterns that differ from normal senders and evaluates the possibility of fraud. The generation AI also builds a system that analyzes email sender information and detects abnormal behavioral patterns based on the past behavioral history. For example, it identifies emails from senders who have been reported to have committed fraud in the past and evaluates the possibility of fraud. As a result, when analyzing email sender information, it can detect abnormal behavioral patterns by referring to the sender's past behavioral history, thereby increasing the possibility of fraudulent emails.

[0087] When analyzing the URL linked from an email, the email analysis unit can evaluate its trustworthiness by referring to the URL's domain information and registration information. For example, the generation AI analyzes the URL linked from an email and evaluates its trustworthiness by referring to the domain information. For example, it identifies links from domains with low reliability and evaluates the possibility of fraud. The generation AI also analyzes the URL linked from an email and evaluates its trustworthiness by referring to the registration information. For example, it identifies URLs with unclear registration information and evaluates the possibility of fraud. The generation AI also analyzes the URL linked from an email and builds a system that evaluates its trustworthiness by combining the domain information and registration information. For example, it identifies domains with low reliability and URLs with unclear registration information and evaluates the possibility of fraud. As a result, when analyzing the URL linked from an email, it can evaluate its trustworthiness by referring to the URL's domain information and registration information, thereby increasing the possibility of it being a fraudulent email.

[0088] The email analysis unit uses an emotion estimation function to estimate the sender's emotions, and can increase the likelihood of fraud if there are abnormal emotional fluctuations. For example, the email analysis unit uses a generation AI to estimate the emotions of an email sender and increase the likelihood of fraud if there are abnormal emotional fluctuations. For example, it evaluates the likelihood of fraud if an email contains strong emotional expressions not seen in normal emails. In addition, a system is constructed in which the generation AI estimates the emotions of an email sender and detects abnormal emotional fluctuations. For example, it evaluates the likelihood of fraud if the sender is feeling anxious or fearful. In addition, a system is constructed in which the generation AI estimates the emotions of an email sender and issues a warning if there are abnormal emotional fluctuations. For example, it evaluates the likelihood of fraud if the sender's emotions change suddenly. In this way, by estimating the emotions of an email sender, it is possible to detect abnormal emotional fluctuations and increase the likelihood of a fraudulent email.

[0089] When analyzing the content of an email, the email analysis unit can simultaneously analyze the content of other communication means to detect signs of fraud. For example, when the generation AI analyzes the content of an email, the email analysis unit simultaneously analyzes the content of other communication means (e.g., SMS and chat) to build a system that detects signs of fraud. For example, if a fraudulent message with the same content is sent via multiple communication means, the possibility of fraud is evaluated. In addition, when the generation AI analyzes the content of an email, it references the content of other communication means to detect signs of fraud. For example, if the content of an email and an SMS match, the possibility of fraud is evaluated. In addition, when the generation AI analyzes the content of an email, it combines and analyzes the content of other communication means to build a system that detects signs of fraud. For example, if the content of an email and a chat match, the possibility of fraud is evaluated. In this way, by simultaneously analyzing the content of other communication means, it is possible to more easily detect signs of fraud.

[0090] When analyzing the content of an email, the email analysis unit can refer to the user's past behavioral history and preferences to detect abnormal content. For example, the email analysis unit uses a generation AI to analyze the user's past behavioral history and detect email content that differs from normal behavioral patterns. For example, it identifies content and formats not found in normal emails and evaluates the possibility of fraud. The generation AI also analyzes the user's preferences and detects email content that differs from normal preferences. For example, it evaluates the possibility of fraud if the email contains links or attachments not found in normal emails. The generation AI also builds a system that combines and analyzes the user's past behavioral history and preferences to detect abnormal email content. For example, it identifies content and formats that differ from normal behavioral patterns and evaluates the possibility of fraud. In this way, by referring to the user's past behavioral history and preferences, it is possible to detect abnormal content and increase the possibility of fraudulent emails.

[0091] The email analysis unit uses an emotion estimation function to monitor the user's emotions in real time when they open an email and can issue a warning if there is an abnormal emotional reaction. For example, the email analysis unit uses a generation AI to monitor the user's emotions in real time when they open an email and issue a warning if there is an abnormal emotional reaction. For example, a warning is displayed if the user feels anxiety or fear. The generation AI also analyzes the user's emotional reaction and issues a warning if there is an abnormal emotional fluctuation. For example, a warning is displayed if there is a strong emotional reaction not seen in normal emails. A system can also be built in which the generation AI monitors the user's emotional reaction in real time and issues a warning if there is an abnormal emotional fluctuation. For example, it analyzes the user's facial expressions and voice and issues a warning if there is an abnormal emotional reaction. This allows the system to monitor the user's emotions in real time when they open an email and issue a warning if there is an abnormal emotional reaction, thereby increasing the possibility of fraud.

[0092] When analyzing the content of a call, the call analysis unit can simultaneously analyze the content of other communication means to detect signs of fraud. For example, when the generation AI analyzes the content of a call, the call analysis unit simultaneously analyzes the content of other communication means (e.g., SMS or chat) to build a system that detects signs of fraud. For example, if the same fraudulent message is sent via multiple communication means, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it references the content of other communication means to detect signs of fraud. For example, if the content of the call matches the content of an SMS, the possibility of fraud is assessed. In addition, when the generation AI analyzes the content of a call, it combines and analyzes the content of other communication means to build a system that detects signs of fraud. For example, if the content of the call matches the content of a chat, the possibility of fraud is assessed. In this way, by simultaneously analyzing the content of other communication means, it is possible to more easily detect signs of fraud.

[0093] When analyzing call content, the call analysis unit can detect abnormal content by referring to the user's past behavioral history and preferences. In the call analysis unit, for example, the generation AI analyzes the user's past behavioral history and detects call content that differs from normal behavioral patterns. For example, it identifies content and formats that are not seen in normal calls and evaluates the possibility of fraud. The generation AI also analyzes the user's preferences and detects call content that differs from normal preferences. For example, it evaluates the possibility of fraud when there are requests or instructions that are not included in normal calls. The generation AI also builds a system that combines and analyzes the user's past behavioral history and preferences to detect abnormal call content. For example, it identifies content and formats that differ from normal behavioral patterns and evaluates the possibility of fraud. In this way, by referring to the user's past behavioral history and preferences, it is possible to detect abnormal content and increase the possibility of fraudulent calls.

[0094] The call analysis unit uses an emotion estimation function to monitor the emotions felt by the user during a call in real time and issue a warning if an abnormal emotional reaction is detected. For example, the call analysis unit uses a generation AI to monitor the emotions felt by the user during a call in real time and issue a warning if an abnormal emotional reaction is detected. For example, a warning is displayed if the user feels anxiety or fear. The generation AI also analyzes the user's emotional reaction and issues a warning if there are abnormal emotional fluctuations. For example, a warning is displayed if there is a strong emotional reaction not seen in a normal call. Furthermore, a system is constructed in which the generation AI monitors the user's emotional reaction in real time and issues a warning if there are abnormal emotional fluctuations. For example, it analyzes the user's facial expressions and voice and issues a warning if there is an abnormal emotional reaction. This allows the system to monitor the emotions felt by the user during a call in real time and issue a warning if there is an abnormal emotional reaction, thereby increasing the possibility of fraud.

[0095] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0096] The fraud detection system may further include a behavior analysis unit that analyzes a user's behavioral history. The behavior analysis unit, for example, analyzes a user's past online behavior and purchase history to detect behavior that differs from normal behavioral patterns. For example, the possibility of fraud is assessed when a high-value purchase that differs from normal purchasing patterns is made. The behavior analysis unit also analyzes a user's past browsing history to detect access to sites that differ from normal browsing patterns. For example, the possibility of fraud is assessed when a suspicious site that differs from normal browsing sites is accessed. Furthermore, the behavior analysis unit analyzes a user's past login history to detect abnormal login patterns. For example, the possibility of fraud is assessed when a login occurs at a time or location that differs from normal login times. In this way, analyzing a user's behavioral history can make it easier to detect signs of fraud.

[0097] The fraud detection system may further include a biometric authentication unit that analyzes the user's biometric information. The biometric authentication unit, for example, analyzes the user's fingerprint or face authentication data and evaluates the possibility of fraud if the data differs from a normal authentication pattern. For example, if a fingerprint different from a normal fingerprint pattern is detected, the possibility of fraud is evaluated. The biometric authentication unit also analyzes the user's voiceprint data and evaluates the possibility of fraud if the data differs from a normal voiceprint pattern. For example, if a voiceprint different from a normal voiceprint is detected, the possibility of fraud is evaluated. The biometric authentication unit also analyzes the user's iris data and evaluates the possibility of fraud if the data differs from a normal iris pattern. For example, if an iris different from a normal iris pattern is detected, the possibility of fraud is evaluated. This makes it easier to detect signs of fraud by analyzing the biometric information.

[0098] The fraud detection system may further include a location information analysis unit that analyzes the user's location information. The location information analysis unit, for example, analyzes the user's current location and evaluates the possibility of fraud if the user is in a location different from their usual range of activity. For example, it evaluates the possibility of fraud if the user is in a country or region different from their usual range of activity. The location information analysis unit also analyzes the user's past movement history and detects abnormal movement patterns. For example, it evaluates the possibility of fraud if there is a sudden movement that differs from the usual movement pattern. Furthermore, the location information analysis unit combines the user's location information with other data and analyzes it to detect abnormal patterns. For example, it evaluates the possibility of fraud if there is a login from a location different from the usual range of activity. In this way, analyzing the location information can make it easier to detect signs of fraud.

[0099] The fraud detection system may further include a device analysis unit that analyzes the user's device information. The device analysis unit, for example, analyzes information about the device used by the user and evaluates the possibility of fraud if it is different from the usual device. For example, it evaluates the possibility of fraud if there is access from a device different from the device normally used. The device analysis unit also analyzes the IP address of the user's device and evaluates the possibility of fraud if it is different from the usual IP address. For example, it evaluates the possibility of fraud if there is access from an IP address different from the usual IP address. The device analysis unit also analyzes the MAC address of the user's device and evaluates the possibility of fraud if it is different from the usual MAC address. For example, it evaluates the possibility of fraud if there is access from a MAC address different from the usual MAC address. In this way, analyzing the device information makes it easier to detect signs of fraud.

[0100] The fraud detection system may further include a social media analysis unit that analyzes the user's social media activity. The social media analysis unit, for example, analyzes the content of the user's posts on social media and evaluates the possibility of fraud if the content differs from the user's normal posting pattern. For example, the social media analysis unit evaluates the possibility of fraud if a suspicious post different from the normal posting content is made. The social media analysis unit also analyzes the user's social media friend list and evaluates the possibility of fraud if the content differs from the normal friend list. For example, the social media analysis unit evaluates the possibility of fraud if a suspicious account not included in the normal friend list is added. The social media analysis unit also analyzes the content of the user's messages on social media and evaluates the possibility of fraud if the content differs from the normal message content. For example, the social media analysis unit evaluates the possibility of fraud if a suspicious message different from the normal message content is sent. In this way, analyzing social media activity can make it easier to detect signs of fraud.

[0101] The fraud detection system may further include an emotion analysis unit that estimates a user's emotions and evaluates the possibility of fraud based on the estimated emotions. The emotion analysis unit, for example, monitors the user's emotions in real time when opening an email and evaluates the possibility of fraud if there is an abnormal emotional reaction. For example, if the user feels anxiety or fear, it evaluates the possibility of fraud. The emotion analysis unit also monitors the user's emotions during a phone call in real time and evaluates the possibility of fraud if there is an abnormal emotional reaction. For example, if the user shows a strong emotional reaction during a phone call, it evaluates the possibility of fraud. Furthermore, the emotion analysis unit monitors the user's emotional fluctuations over the long term and evaluates the possibility of fraud if they differ from the normal emotional pattern. For example, it evaluates the possibility of fraud if there is a sudden emotional change that differs from the normal emotional pattern. In this way, analyzing emotions can make it easier to detect signs of fraud.

[0102] The fraud detection system may further include a warning emotion analysis unit that estimates a user's emotions and issues a warning based on the estimated emotions. The warning emotion analysis unit, for example, monitors the user's emotions in real time when opening an email and issues a warning if an abnormal emotional reaction is detected. For example, a warning is displayed if the user feels anxiety or fear. The warning emotion analysis unit also monitors the user's emotions during a call in real time and issues a warning if an abnormal emotional reaction is detected. For example, a warning is displayed if the user shows a strong emotional reaction during a call. The warning emotion analysis unit also monitors the user's emotional fluctuations over the long term and issues a warning if the emotional pattern differs from the normal emotional pattern. For example, a warning is displayed if there is a sudden emotional change that differs from the normal emotional pattern. In this way, the possibility of fraud can be increased by analyzing emotions and issuing a warning if an abnormal emotional reaction is detected.

[0103] The fraud detection system may further include a learning emotion analysis unit that estimates a user's emotions and learns fraud patterns based on the estimated emotions. The learning emotion analysis unit, for example, analyzes the emotions felt when a user opens an email to learn fraud patterns. For example, it may learn patterns of emails in which the user felt anxiety or fear, and use this information to help detect future fraudulent emails. The learning emotion analysis unit may also analyze the emotions felt by a user during a phone call to learn fraud patterns. For example, it may learn patterns of calls in which the user showed strong emotional reactions, and use this information to help detect future fraudulent calls. The learning emotion analysis unit may also analyze a user's emotional fluctuations over the long term to learn fraud patterns. For example, it may learn patterns of sudden emotional fluctuations that differ from normal emotional patterns, and use this information to help detect future fraud. In this way, analyzing emotions and learning fraud patterns can make it easier to detect signs of fraud.

[0104] The fraud detection system may further include a risk emotion analysis unit that estimates the user's emotions and assesses the risk of fraud based on the estimated emotions. The risk emotion analysis unit, for example, analyzes the emotions felt when the user opens an email and assesses the risk of fraud. For example, if the user feels anxiety or fear, the risk of fraud is assessed as high. The risk emotion analysis unit also analyzes the emotions felt by the user during a call and assesses the risk of fraud. For example, if the user shows a strong emotional reaction during a call, the risk of fraud is assessed as high. Furthermore, the risk emotion analysis unit analyzes the user's emotional fluctuations over the long term and assesses the risk of fraud. For example, if there is a sudden emotional fluctuation that differs from normal emotional patterns, the risk of fraud is assessed as high. In this way, analyzing emotions and assessing the risk of fraud makes it easier to detect signs of fraud.

[0105] The fraud detection system may further include a preventive emotion analysis unit that estimates a user's emotions and provides fraud prevention measures based on the estimated emotions. The preventive emotion analysis unit, for example, analyzes the emotions a user feels when opening an email and provides fraud prevention measures. For example, if the user feels anxiety or fear, a preventive measure is provided that notifies the user of a possible fraud. The preventive emotion analysis unit also analyzes the emotions a user feels during a call and provides fraud prevention measures. For example, if the user shows a strong emotional reaction during a call, a preventive measure is provided that notifies the user of a possible fraud. The preventive emotion analysis unit also analyzes the user's emotional fluctuations over the long term and provides fraud prevention measures. For example, if there is a sudden emotional fluctuation that differs from a normal emotional pattern, a preventive measure is provided that notifies the user of a possible fraud. In this way, analyzing emotions and providing fraud prevention measures makes it easier to detect signs of fraud.

[0106] The processing flow of the second embodiment will be briefly explained below.

[0107] Step 1: The email analysis unit analyzes the email received by the user. The generation AI analyzes the email content, sender information, linked URLs, etc. to determine whether there is a possibility of fraud. For example, it can use natural language processing technology to analyze the email content and detect signs of fraud. It can also verify the sender's information and evaluate their trustworthiness. It can also analyze linked URLs and evaluate their trustworthiness. Step 2: The call analysis unit analyzes the content of the call received by the user. The generation AI converts the content of the call into text using voice recognition technology and analyzes the text. For example, it can analyze the content of the call using natural language processing technology to detect signs of fraud. It can also analyze the audio data of the call and detect abnormal patterns. It can also estimate the user's emotions using an emotion estimation function. Step 3: The warning notification unit issues a warning to the user if a possible fraud is detected. For example, if a fraudulent email is detected, a warning message is displayed in the user's email app. Also, if a fraudulent phone call is detected, a warning sound is played during the call or a warning message is displayed after the call ends. Step 4: The database updater records the detected fraud information in a database and shares it with other users. For example, if a specific email address or phone number is determined to be related to fraud, that information is added to the database.

[0108] 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.

[0109] 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> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). 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 speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. 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. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0110] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0111] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0112] 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0113] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0114] 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.

[0115] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0116] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0117] 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.

[0118] 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.

[0119] 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.

[0120] 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. The identification processing unit 290 can estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0121] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart glasses 214 may have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59.

[0122] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0123] 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.

[0124] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0125] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0126] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0127] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0128] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0129] 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.

[0130] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0131] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0132] 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.

[0133] 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.

[0134] 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.

[0135] 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. The identification processing unit 290 can estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0136] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 may also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59.

[0137] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0138] 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.

[0139] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0140] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0141] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0142] 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0143] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0144] 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.

[0145] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0146] 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 image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0147] 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.

[0148] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the 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.

[0149] 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.

[0150] 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.

[0151] 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. The identification processing unit 290 can estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0152] In the robot 414, the processor 46 performs the identification process. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 may have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59.

[0153] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0154] 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.

[0155] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0156] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0157] 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.

[0158] FIG. 9 illustrates 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 behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions 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.

[0159] 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.

[0160] 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).

[0161] 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 expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, 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 expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0162] 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."

[0163] 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.

[0164] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0165] 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.

[0166] 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.

[0167] 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.

[0168] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, 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. A processor also includes 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.

[0169] The hardware resource that executes the specific process 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 process may be a single processor.

[0170] 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.

[0171] 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.

[0172] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0173] 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.

[0174] 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. [Explanation of symbols]

[0175] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. A system that uses generative AI to detect signs of fraud and issue a warning to users, an email analysis unit that analyzes emails received by the user; a call analysis unit that analyzes the content of a call received by the user; a warning notification unit that issues a warning to the user when a possible fraud is detected; It has a database updater that records detected fraud information in a database and shares it with other users. A system characterized by:

2. The email analysis unit When analyzing the content of an email, the content of any images or attachments may also be analyzed to detect signs of said fraud.

2. The system of claim 1.

3. The call analysis unit Analyze call start times or durations to detect unusual patterns 2. The system of claim 1.

4. The database update unit Recording the detected fraud information in the database and sharing it with the other users.

2. The system of claim 1.

5. The email analysis unit The user's emotions are estimated from the content of the email, and if there are abnormal emotional fluctuations, the possibility of fraud is increased.

2. The system of claim 1.

Images