system

Abstract

A system is provided.SOLUTION: A system includes means for a user to input past interactions in an SNS or a matching application, means for transmitting input data to a server, means for the server to input data received from the user to an AI analysis system, means for the AI analysis system to analyze the data and extract elements based on patterns of fraudsters, means for the server to extract patterns of fraudsters from a database based on analysis results, means for the server to evaluate fraud risk based on the analysis results and generate a warning message, and means for notifying the user of the generated warning message.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In recent years, fraudulent activities via social media and dating apps have been increasing, particularly those involving marriage or investment offers. These fraudulent activities are sophisticated and many of them exploit the victim's normalcy bias, making it difficult to prevent them from occurring. Victims often have difficulty detecting the signs of fraud, resulting in financial losses and psychological shock. To solve this problem, a system is needed that can detect fraud risks in advance and issue warnings when users use social media or dating apps. [Means for solving the problem]

[0005] The present invention provides a means for users to input their past interactions on social media or dating apps and send that data to a server. The server then inputs the data received from the user into an AI analysis system for analysis. The AI ​​analysis system extracts elements based on fraudster patterns and, based on the analysis results, extracts fraudster patterns from a database. The server then evaluates fraud risk based on the analysis results, generates a warning message, and notifies the user of the message. The AI ​​analysis system also has the ability to receive user feedback and update the model. Furthermore, the analyzed data is securely transmitted using encryption protocols such as SSL. In this way, users can understand fraud risks in advance and take appropriate action.

[0006] "User" refers to an individual who interacts with others on social media or matching apps.

[0007] "SNS" refers to all social networking services on the Internet.

[0008] A "matching app" refers to software developed for the purpose of connecting individuals.

[0009] "Server" refers to a computer system that receives, analyzes, and processes data from users.

[0010] "AI analysis system" refers to a system that uses artificial intelligence technology to analyze data and detect signs of fraud.

[0011] "Database" refers to an information collection system that stores fraudster patterns and past cases and allows for searching and retrieving them as needed.

[0012] "Fraudster patterns" refer to the common tendencies of behavior and speech that fraudsters have.

[0013] "Analysis results" means the results of data analysis obtained by the AI ​​analysis system.

[0014] "Fraud risk" refers to the degree to which an interaction may involve fraud.

[0015] "Warning message" refers to a notice to alert users to possible fraud and to draw their attention to it.

[0016] "Feedback" refers to opinions and evaluations provided by users, and is information that is used to improve AI models.

[0017] "SSL and other encryption protocols" refers to encryption technologies used to securely communicate data.

[0018] "Model updating" refers to the process by which an AI analytics system incorporates new data and feedback to improve its performance. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] 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.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] 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.

[0024] 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.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] 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."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] 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.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] 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.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, and evaluates the possibility of fraud. Specific embodiments are described below.

[0041] System Overview

[0042] The system mainly consists of the following components:

[0043] 1. User Device

[0044] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[0045] 2. Server

[0046] The server is the central computer system that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[0047] 3. AI Analysis System

[0048] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[0049] 4. Database

[0050] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[0051] Program processing explanation

[0052] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0053] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[0054] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[0055] Based on the analysis results, the server extracts fraudster patterns from the database and assesses the fraud risk, for example, calculating that the probability of fraud is 80%.

[0056] The server generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check for more information." and is sent to the user's device.

[0057] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model.

[0058] Specific examples

[0059] Example 1: Social media interactions

[0060] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0061] 2. The user terminal sends this text to the server.

[0062] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0063] 4. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[0064] 5. The server creates a warning message saying, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[0065] Example 2: Interactions on a dating app

[0066] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[0067] 2. The user terminal sends this text to the server.

[0068] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0069] 4. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[0070] 5. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[0071] In this way, users can analyze the true meaning of interactions on social media and dating apps and prevent themselves from falling victim to fraud.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0075] Step 2:

[0076] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[0077] Step 3:

[0078] The server receives the data sent from the user terminal and confirms that the data has been received.

[0079] Step 4:

[0080] Before the server can pass the received data to the AI ​​analysis system, it cleanses the text data and formats it in a way that makes it easier to analyze, for example by removing unnecessary spaces and special characters.

[0081] Step 5:

[0082] An AI analysis system analyzes incoming data and extracts elements based on scammer patterns, specifically identifying phrases that indicate fraudulent activity, such as "please lend me some money."

[0083] Step 6:

[0084] Based on the results of the AI ​​analysis system, the server extracts fraudster patterns from a database that records past fraud cases and typical patterns of fraudulent behavior.

[0085] Step 7:

[0086] The server will evaluate multiple factors (e.g., phrases used, frequency of interactions, inconsistencies, etc.) and calculate a fraud risk as a percentage. For example, it may determine that the fraud risk is 80%.

[0087] Step 8:

[0088] The server generates a warning message based on the fraud risk assessment result, such as "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[0089] Step 9:

[0090] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[0091] Step 10:

[0092] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[0093] Step 11:

[0094] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[0095] Step 12:

[0096] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model in the future.

[0097] This is the specific processing flow of this system, which allows users to detect in advance the risk of becoming a victim of fraud through interactions on social media or dating apps, and respond appropriately.

[0098] Example 1

[0099] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0100] In modern society, fraudulent activities involving the misuse of communication applications are becoming more common. Users of social networking sites and dating apps are particularly susceptible to the sophisticated tactics of fraudsters. In addition to standard security measures, there is a demand for systems that can accurately detect potential fraudulent activity and quickly warn users. However, conventional systems have difficulty efficiently detecting signs of fraud and incorporating user feedback to improve accuracy. The purpose of this invention is to solve these problems.

[0101] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0102] In this invention, the server includes: a means for a user to input past interactions in a communication application; a means for transmitting the input data to an information processing device; a means for the information processing device to input the data received from the user to a machine learning model; a means for the machine learning model to analyze the data and extract elements based on fraudulent activity patterns; a means for the information processing device to extract fraudulent activity patterns from a database based on the analysis results; a means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; and a means for notifying the user of the generated warning message. This enables the system to detect possible fraudulent activity with high accuracy and quickly issue a warning to the user. Furthermore, by adding a means for updating the machine learning model based on user feedback, the system can continuously improve its accuracy. As a result, users can use communication applications with peace of mind.

[0103] "User" refers to an individual or corporation that uses this system.

[0104] "Communication applications" refers to all software that allows messages and information to be exchanged over the Internet, such as social networking sites and dating apps.

[0105] "Past interactions" refers to the history of messages and chats that a user has had on a communication application.

[0106] "Information processing device" refers to a general term for a computer system that receives, analyzes, stores, and notifies data.

[0107] A "machine learning model" refers to an algorithm that uses artificial intelligence to learn patterns from past data and analyze and predict new data.

[0108] "Fraud patterns" refer to common behavioral characteristics or phrases extracted from past instances of fraud or malicious behavior.

[0109] "Analysis results" refer to the indicators and elements of fraudulent activity extracted by the machine learning model from the input data.

[0110] A "database" refers to a system that stores information necessary for analysis, such as patterns of fraudulent activity and past cases.

[0111] "Fraud risk" refers to the possibility of fraud or malicious intent calculated based on the analysis results.

[0112] The "warning message" refers to information sent to the user to call their attention based on the analysis results.

[0113] "Feedback" refers to the opinions and evaluations provided by users regarding the system's analysis results.

[0114] MODE FOR CARRYING OUT THE INVENTION

[0115] The present invention is a system that allows users to input their past interactions on social media or dating apps, analyzes that data with an AI analysis system, evaluates the possibility of fraud, and issues a warning to the user. The program processing of this system is explained in detail below.

[0116] System configuration

[0117] This invention mainly consists of the following hardware and software:

[0118] 1. User Device

[0119] A device that allows a user to input past interactions with a communication application, including a smartphone or computer.

[0120] 2. Server

[0121] The server is the central information processing device that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[0122] 3. AI Analysis System

[0123] The AI ​​analysis system analyzes data based on fraud patterns and extracts elements of fraudulent activity. It uses machine learning algorithms to enable highly accurate data analysis. For example, it uses machine learning frameworks such as TENSORFLOW (registered trademark) and PyTorch.

[0124] 4. Database

[0125] It is an information collection system that stores fraud patterns and past cases. The server refers to this database to assess fraud risk.

[0126] Program processing overview

[0127] First, the user inputs a past conversation in a communication application in text format. For example, the user might input something like, "Hello! Would you like to meet sometime? Can you lend me a little money first?" into the system.

[0128] The user terminal then sends the entered data to the server, using an encryption protocol such as SSL to ensure the security of the communication.

[0129] The server passes the received data to an AI analysis system, which extracts elements that indicate fraudulent activity. The AI ​​analysis system learns from past cases of fraud and identifies and analyzes phrases such as "lend me some money."

[0130] Based on the analysis results, the server extracts fraudulent patterns from the database and assesses the fraud risk. For example, if the probability of fraud is calculated to be 80%, it generates a specific warning message to inform the user of the risk.

[0131] The generated warning message may say something like, "This person is likely to be a fraud. Avoid lending money to them and check for more information," and is sent from the server to the user's terminal.

[0132] Users can provide feedback on the analysis results. This feedback is sent to the server and used to improve the AI ​​model. For example, the feedback could be, "This warning was appropriate."

[0133] Specific examples

[0134] Example 1: Social media interactions

[0135] 1. The user enters the chat history from a social networking site into the system: "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[0136] 2. The user terminal sends this text to the server.

[0137] 3. The server passes the received data to the AI ​​analysis system.

[0138] 4. The server's AI analysis system determines that the phrase "lend me money" is 80% likely to be a scam.

[0139] 5. Based on the results, the server generates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and notifies the user.

[0140] 6. The user sends feedback saying, "This warning was appropriate."

[0141] 7. The server stores the feedback in a database and uses it for future analysis.

[0142] Example 2: Interactions on a dating app

[0143] 1. The user enters the chat history from the dating app into the system: "I want to get married soon! There's an investment opportunity. Are you interested?"

[0144] 2. The user device sends this data to the server.

[0145] 3. The server passes the received data to the AI ​​analysis system.

[0146] 4. The server's AI analysis system detects phrases related to "investment opportunities" and rates the likelihood of fraud at 90%.

[0147] 5. Based on the high fraud risk, the server generates a warning message stating, "This person is very likely to be a fraud. Please be especially careful with investment proposals and check the detailed information." and notifies the user.

[0148] 6. The user sends feedback saying, "This warning was helpful."

[0149] 7. The server stores the feedback in a database and uses it to improve the AI ​​model.

[0150] In this way, the system of the present invention can analyze the true intent of users' interactions in communication applications and prevent fraud before it occurs. Furthermore, by incorporating user feedback, the system's analysis accuracy can be continuously improved.

[0151] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0152] Step 1:

[0153] The user inputs past conversations in a communication application in text format. For example, the user copies and pastes the chat content from an SNS such as "Hello! Let's meet sometime. Can you lend me a little money first?" into the system. The input text is temporarily stored on the user's device.

[0154] Step 2:

[0155] The user terminal sends the entered text data to the server. At this time, security is ensured when sending data using encryption protocols such as SSL. Specifically, when the user presses the "Send" button, the text data is encrypted and sent to the server via the Internet. The input is the text data entered by the user, and the output is the encrypted data received by the server.

[0156] Step 3:

[0157] The server receives the data sent from the user terminal. The received data is decrypted and restored to the original text data. Specifically, the decryption program on the server decrypts the encrypted data according to the SSL protocol to obtain plain text data. The input is encrypted text data, and the output is decrypted text data.

[0158] Step 4:

[0159] The server passes the decrypted text data to the AI ​​analysis system. The passed data is added to a processing queue within the AI ​​analysis system. Specifically, the server converts the text data into a format suitable for the machine learning model and sends it to the analysis system. The input is the decrypted text data, and the output is the data input to the machine learning model.

[0160] Step 5:

[0161] The AI ​​analysis system analyzes the data and extracts elements that indicate fraudulent activity. For example, it identifies the phrase "lend me some money" and extracts corresponding patterns. The input is the text data to be analyzed, and the output is the extracted elements of fraudulent activity.

[0162] Step 6:

[0163] The server then references the database based on the analysis results and compares them with past fraudulent activity patterns. Specifically, the server executes a database query to search for relevant past cases. The input is the extracted fraudulent activity elements, and the output is the corresponding fraudulent activity patterns retrieved from the database.

[0164] Step 7:

[0165] The server assesses the fraud risk based on the fraudulent activity patterns and analysis results. This assessment is converted into a numerical value according to an algorithm, for example, the probability of fraud is calculated as 80%. Specifically, the server executes the risk assessment algorithm and calculates the fraud risk score. The input is the fraudulent activity patterns and analysis results obtained from the database, and the output is the fraud risk score.

[0166] Step 8:

[0167] The server generates a warning message based on the fraud risk. For example, it creates a warning message that reads, "This person is likely to be a fraudster. Avoid lending money to them and check for more information." The input is the fraud risk score, and the output is the generated warning message.

[0168] Step 9:

[0169] The server sends the generated warning message to the user terminal. The sent message is displayed on the user terminal. Specifically, the server sends a message to the user terminal using a message transmission protocol, and the user terminal displays the message on its screen. The input is the generated warning message, and the output is the warning message displayed on the user terminal.

[0170] Step 10:

[0171] The user provides feedback on the analysis results. This feedback is sent from the user terminal to the server. For example, the user may input "This warning was appropriate." The input is the feedback on the analysis results, and the output is the feedback data sent to the server.

[0172] Step 11:

[0173] The server receives user feedback and updates the machine learning model of the AI ​​analysis system. Specifically, the server passes the feedback data to the AI ​​analysis system and retrains the machine learning algorithm. The input is the feedback data, and the output is the updated machine learning model.

[0174] (Application example 1)

[0175] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0176] Users of communication platforms such as social networking sites and dating apps are at increased risk of becoming victims of fraud. Under these circumstances, there is a need for a method that automatically and accurately analyzes users' risk of fraud based on their past interactions and prevents fraud from occurring. However, with conventional methods, it is difficult for users to determine the risk themselves, which often results in delayed response. The present invention aims to solve this problem.

[0177] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0178] In this invention, the server includes: means for a user to input past interactions on the communication platform; means for transmitting the input data; means for the information processing device to input the data received from the user to a knowledge base analysis system; means for the knowledge base analysis system to analyze the data and extract elements based on fraud patterns; means for the information processing device to extract fraud patterns from the information accumulation system based on the analysis results; means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; means for notifying the user of the generated warning message; a mobile terminal having a user interface for copying and pasting interactions into an application; and a data transmission module having a protocol for encrypting and transmitting the input data. This enables a user to easily input past interactions and have their risks quickly and accurately evaluated.

[0179] "User" refers to any individual or organization using the communication platform.

[0180] A "communication platform" refers to an online service that allows users to exchange messages with each other, such as social networking sites or matching apps.

[0181] "Interaction" refers to the sending and receiving of messages between users on the communication platform.

[0182] "Input means" refers to a method or device that allows a user to provide the content of an interaction to the system in text form.

[0183] "Information processing device" refers to a computer device that receives, processes, and transmits data.

[0184] "Knowledge-based analysis system" refers to an artificial intelligence system for analyzing data based on patterns of fraudulent activity.

[0185] "Means for extracting elements" refers to the methods and devices that the knowledge-based analysis system uses to find features and patterns related to fraudulent activity in the data.

[0186] An "information collection system" refers to a database that accumulates and manages fraud patterns and past cases.

[0187] "Warning message" refers to a message intended to inform users of the risk of fraud.

[0188] "User interface" refers to the screens and input devices through which a user interacts with a system and inputs data.

[0189] "Mobile terminal" refers to an electronic device that a user can carry with them, such as a smartphone or tablet.

[0190] An "encryption and transmission protocol" refers to a method or communication protocol for encrypting and transmitting data to ensure secure data communication.

[0191] The system of the present invention allows users to input their past interactions on a communication platform, analyzes their hidden intentions using a knowledge-based analysis system, and assesses the risk of fraud. Specific embodiments are described below.

[0192] System Overview

[0193] The system mainly consists of the following components:

[0194] 1. User Device

[0195] A mobile device for inputting past interactions on a communication platform via a user interface. Users use a smartphone or tablet to input the relevant text into the system.

[0196] 2. Information Processing Device

[0197] It is a central computer system that receives data and executes the knowledge-based analysis system. The information processing device also manages the information collection system and notifies the results.

[0198] 3. Knowledge-based analysis system

[0199] The system analyzes data based on fraud patterns and extracts relevant elements. The knowledge-based analysis system uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[0200] 4. Information Collection System

[0201] This is a database for storing fraud patterns and past cases. The information processing device refers to this information accumulation system to assess fraud risk.

[0202] Processing Step Description

[0203] First, the user inputs a past conversation on the communication platform in text format, for example, "Hello! Would you like to meet up sometime? Can you lend me a little money first?" and pastes it into the application.

[0204] Next, the user terminal transmits the input data to the information processing device. This data transmission uses an encryption protocol such as SSL to ensure the security of the communication.

[0205] The information processing device passes the received data to a knowledge-based analysis system, which extracts elements that indicate fraud. The knowledge-based analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[0206] Based on the analysis results, the information processing device extracts fraud patterns from the information collection system and evaluates the fraud risk. For example, it calculates that the probability of fraud is 80%.

[0207] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and is sent to the user terminal.

[0208] Additionally, users can provide feedback on the analysis results, which is sent to the information processing device and used to improve the knowledge-based analysis system.

[0209] Specific examples

[0210] Example 1: Analysis of social media interactions

[0211] Type: "Hi! Want to meet up sometime? Can I borrow a little money first?"

[0212] Analysis result: 80% fraud risk

[0213] Notification message: "This person is likely a scammer. Avoid lending money and check for more information."

[0214] Example 2: Analysis of interactions on a dating app

[0215] Type: "I'd love to get married soon! I have an investment opportunity, would you be interested?"

[0216] Analysis result: 90% fraud risk

[0217] Notification message: "This person is very likely a scam. Be especially careful with investment proposals and check the full details."

[0218] Prompt Sentence Examples

[0219] "Analyze the following exchange and assess the risk of fraud: 'Hi! Would you like to meet up sometime? Can I borrow a little money first?'"

[0220] Analyze the following exchange and assess its risk of fraud: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[0221] In this way, users can analyze the true meaning of interactions on the communication platform and prevent themselves from falling victim to fraud.

[0222] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0223] Step 1:

[0224] A user inputs past interactions on a communication platform in text format. The user inputs (copy and paste) the text via the user interface of a smartphone or tablet, and prepares this text for subsequent analysis. The input includes content such as "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[0225] Step 2:

[0226] The user terminal transmits the input data to the information processing device. An encryption protocol such as SSL is used for data transmission to ensure communication security. The input in this step is the text entered by the user, and the output is the transmission of encrypted data.

[0227] Step 3:

[0228] The information processing device passes the received data to a knowledge-based analysis system. This data is decrypted in its original text form and provided to the knowledge-based analysis system for analysis. The input is the decrypted content of the encrypted data, and the output is analyzable text data.

[0229] Step 4:

[0230] A knowledge-based analysis system analyzes the data and extracts indicators of fraud. It uses a generative AI model trained on past fraud cases to identify phrases such as "lend me some money." The input is analyzable text data, and the output is a fraud risk score and indicators of fraud.

[0231] Step 5:

[0232] The data processor extracts fraud patterns from the data collection system based on the analysis results and assesses the fraud risk. The data collection system stores past fraud patterns and cases, and compares them with the current input data. The input is a fraud risk score and fraud indicator elements, and the output is a final fraud risk assessment.

[0233] Step 6:

[0234] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. The generated warning message includes content such as "This person is likely to be a fraudster. Avoid lending money and check the detailed information." The input is the fraud risk assessment, and the output is a warning message to the user.

[0235] Step 7:

[0236] The user provides feedback on the analysis results. The feedback is sent back to the information processing system, and the knowledge-based analysis system updates the model based on this feedback. The input in this step is the user's feedback, and the output is the updated analytical model.

[0237] This process allows users to quickly and automatically analyze whether their interactions on the communication platform pose a risk of fraud and receive appropriate warnings.

[0238] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0239] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, evaluates the possibility of fraud, and provides more accurate warnings by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments are described below.

[0240] System Overview

[0241] The system mainly consists of the following components:

[0242] 1. User Device

[0243] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[0244] 2. Server

[0245] It is the central computer system that receives data and runs the AI ​​analysis system and emotion engine. The server also manages the database and notifies the results.

[0246] 3. AI Analysis System

[0247] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[0248] 4. Emotion Engine

[0249] Analyzing emotions from user input data and classifying them into positive, negative, and neutral categories, the emotion engine provides emotional information to the AI ​​analysis system, which then adds it to the fraud risk assessment.

[0250] 5. Database

[0251] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[0252] Program processing explanation

[0253] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0254] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[0255] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[0256] At the same time, the server also passes the received data to the emotion engine, which analyzes the user's emotions. The emotion engine analyzes the emotions in the text and classifies them as positive, negative, or neutral.

[0257] Based on the analysis results, the server extracts fraudster patterns from the database and evaluates the fraud risk. Emotional information from the emotion engine is also added to the fraud risk assessment. For example, the probability of fraud is 80% and the user's emotion is determined to be negative.

[0258] The server generates a warning message based on the fraud risk assessment results and emotional information and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and sent to the user's terminal.

[0259] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model and emotion engine.

[0260] Specific examples

[0261] Example 1: Social media interactions

[0262] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0263] 2. The user terminal sends this text to the server.

[0264] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0265] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as negative.

[0266] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[0267] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[0268] Example 2: Interactions on a dating app

[0269] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[0270] 2. The user terminal sends this text to the server.

[0271] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0272] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as neutral.

[0273] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[0274] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[0275] In this way, users can analyze their interactions on social media and dating apps to understand their emotional state as well as the risk of fraud, enabling more accurate risk assessment and appropriate responses.

[0276] The processing flow will be explained below.

[0277] Step 1:

[0278] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0279] Step 2:

[0280] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[0281] Step 3:

[0282] The server receives the data sent from the user terminal and confirms that the data has been received.

[0283] Step 4:

[0284] The server cleanses the incoming data and formats it in a way that makes it easier to parse, for example by removing extra whitespace and special characters.

[0285] Step 5:

[0286] The server passes the cleansed data to the AI ​​analysis system, which then analyzes the data and extracts elements that indicate fraudulent activity, such as requests to lend money.

[0287] Step 6:

[0288] The server then passes the same cleansed data to a sentiment engine, which analyzes the data and identifies sentiments such as positive, negative, or neutral.

[0289] Step 7:

[0290] The emotion engine returns the analysis results to the server and determines whether the user has a negative emotion. For example, the emotion engine classifies the emotion as "negative."

[0291] Step 8:

[0292] The server extracts fraudster patterns from the database based on the results of the AI ​​analysis system and the emotion engine, and matches patterns that contain elements that indicate fraudulent activity.

[0293] Step 9:

[0294] The server comprehensively evaluates the fraud risk based on the analysis results and emotional information. For example, if the fraud rate is 80% and the user's emotional state is negative,

[0295] Step 10:

[0296] The server generates a warning message based on the fraud risk assessment results and emotional information. For example, it creates a message saying, "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[0297] Step 11:

[0298] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[0299] Step 12:

[0300] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[0301] Step 13:

[0302] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[0303] Step 14:

[0304] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model and emotion engine in the future.

[0305] This is the specific processing flow of this system. As a result, users can detect fraudulent behavior early through their interactions on social media or dating apps, as well as recognize their own emotional state and take appropriate action.

[0306] Example 2

[0307] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0308] In conventional social networking services and dating applications, there was no system that effectively assessed and warned users about their risk of being scammed. As a result, users were more likely to fall prey to scammers' sophisticated tactics, increasing the risk of financial loss. In addition, the accuracy of warnings was low because the system did not take into account the user's emotional state.

[0309] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to input past interactions on a social networking service or a matching application; a means for transmitting the input data; a means for the processing device to input data received from the user to an AI analysis system; a means for the AI ​​analysis system to analyze the data and extract elements based on fraud patterns; a means for the processing device to extract fraud patterns from the data management system based on the analysis results; a means for the processing device to evaluate fraud risk based on the analysis results and generate a warning message; a means for notifying the user of the generated warning message; and a means for analyzing emotions in the input data and adding emotion-based information to the fraud risk evaluation. This allows users to quickly and accurately evaluate the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[0310] "User" refers to an individual who uses a social networking service or dating application.

[0311] "Social networking service" refers to an online platform that enables users to communicate with other individuals.

[0312] "Matching Application" means a mobile or web application that enables users to meet and connect with other individuals.

[0313] "Past interactions" refers to the history of messages exchanged by a user on social networking services or matching applications.

[0314] "Means for input" refers to an interface or device that allows a user to send past interactions to the system in text form.

[0315] "Processing device" refers to a central computer system for receiving, analyzing, and evaluating input data.

[0316] "Artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze data and detect patterns of fraudulent activity.

[0317] A "data management system" refers to an information storage system that stores past fraud cases and patterns and references them as needed.

[0318] "Fraud" refers to any dishonest activity that attempts to deceive users into giving them money or personal information.

[0319] "Elements" refer to the characteristics extracted by artificial intelligence analysis systems to identify patterns of fraudulent activity.

[0320] "Analysis results" refers to the conclusions and data derived by the AI ​​analysis system through data analysis.

[0321] "Fraud risk" refers to the degree of likelihood that the input data represents fraudulent activity.

[0322] "Warning Message" refers to a message containing a warning provided to a user based on a fraud risk assessment.

[0323] "Means of notification" refers to the communication means or interface used to convey a warning message to the user.

[0324] "Means for analyzing emotions" refers to systems or algorithms for extracting emotional information from text data and analyzing that information.

[0325] "Emotion-based information" refers to data that represents a user's emotional state and is used in fraud risk assessment.

[0326] This invention provides a system that inputs past interactions a user has had on social networking services or matching applications, analyzes the data using an AI analysis system and a sentiment analysis engine, and assesses the risk of fraud. This system is comprised of a user terminal, a server, an AI analysis system, a sentiment analysis engine, and a database.

[0327] Hardware and software used

[0328] User device: The smartphone or computer used by the user to enter data and receive results. Examples: iPhone®, ANDROID® device, Windows PC, Mac.

[0329] Server: A high-performance computer system for receiving data and performing analytical processing. Examples include AWS® EC2 and Google® Cloud Compute Engine.

[0330] Communication Protocol: The secure communication method used to send and receive data. For example: SSL / TLS.

[0331] Artificial intelligence analysis systems: Use machine learning algorithms that learn fraud patterns. Examples: Scikit-learn, TensorFlow.

[0332] Sentiment analysis engine: Uses algorithms that analyze user sentiment. Example: IBM Watson® Natural Language Understanding.

[0333] Database: An information storage system for storing past fraud cases and patterns. Examples include MySQL (registered trademark), PostgreSQL.

[0334] System Overview

[0335] 1. User enters data:

[0336] The user inputs past interactions in text format, for example, "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[0337] 2. Data transmission:

[0338] The user terminal sends the entered data to the server using an encryption protocol (such as SSL / TLS).

[0339] 3. Data Receipt and Analysis:

[0340] The server receives the data and first analyzes it with an artificial intelligence analysis system, which uses machine learning algorithms that have learned fraudulent patterns to extract elements, such as phrases like "lend me some money."

[0341] 4. Emotion analysis:

[0342] The data is passed to a sentiment analysis engine to analyze the user's emotional state. The text is classified as positive, negative, or neutral. For example, if the user's message is offensive, it will be classified as negative.

[0343] 5. Fraud risk assessment and warning message generation:

[0344] The server integrates the results of the AI ​​analysis system and the sentiment analysis engine, and compares them with past fraud patterns in the database to assess the fraud risk. For example, if the fraud risk is determined to be high (80%), a warning message is generated based on that risk.

[0345] 6. Warning message notification:

[0346] The server generates a warning message and sends it to the user's device, which displays the message "This person is likely a fraud. Avoid lending money to them and check for more information."

[0347] 7. Gathering Feedback:

[0348] Users provide feedback on the analysis results, which is sent to the server and used to improve the artificial intelligence model and sentiment analysis engine.

[0349] Specific examples

[0350] Example 1: Social media interactions

[0351] 1. Users copy social media conversations and enter them into the system.

[0352] For example: "Hi! Would you like to meet up sometime? Can I borrow a little money first?"

[0353] 2. The user terminal sends this text to the server.

[0354] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[0355] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as negative.

[0356] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[0357] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[0358] Example 2: Interactions on a dating app

[0359] 1. The user copies the conversations they have had on the dating app and enters them into the system.

[0360] Example: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[0361] 2. The user terminal sends this text to the server.

[0362] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[0363] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as neutral.

[0364] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[0365] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[0366] This allows users to quickly and accurately assess the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[0367] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0368] Step 1:

[0369] The user inputs past interactions in text format on social networking services or dating applications. Specifically, the user inputs past messages by copying and pasting them into the application screen on their smartphone or computer. An example of input is the message "Hello! Would you like to meet sometime? Can you lend me a little money first?" The input data is saved in text format.

[0370] Step 2:

[0371] The user terminal sends the entered text data to the server. At this time, an encryption protocol such as SSL / TLS is used to ensure the security of the communication. The input is text data, and the output is encrypted data. In concrete terms, the application on the user terminal issues a send request to the server, and a progress bar is displayed.

[0372] Step 3:

[0373] The server passes the received data to an artificial intelligence analysis system, which extracts elements that indicate fraud. The data input is encrypted text data, and the output contains elements related to fraud patterns (e.g., keywords such as "money" and "lend me"). This processing uses machine learning algorithms (e.g., Scikit-learn, TensorFlow). Specifically, the server records a "receiving data" log and displays an analysis progress bar in percentage.

[0374] Step 4:

[0375] At the same time, the server passes the received data to a sentiment analysis engine to analyze the user's sentiment. The data input is the original text data, and the output is the sentiment classification result (positive, negative, neutral). A sentiment analysis algorithm (e.g., IBM Watson Natural Language Understanding) is used for sentiment analysis. Specifically, the server records a log of "sentiment analysis in progress" and displays the progress of sentiment analysis in percentage.

[0376] Step 5:

[0377] The server integrates the results of the AI ​​analysis system and sentiment analysis, and compares them with past fraud patterns in the database to assess the fraud risk. The input is the elements of the fraud pattern and the sentiment classification results, and the output is the fraud risk assessment result (e.g., fraud risk 80%). The server generates a warning message based on the analysis results. Specifically, the server records a log of "fraud risk assessment in progress" and displays the fraud risk value.

[0378] Step 6:

[0379] The server generates a warning message and sends it to the user's device. The input is the fraud risk assessment result, and the output is a warning message (e.g., "This person is likely to be a fraudster. Avoid lending money and check for more information."). Specifically, the server issues a message sending request to the user's device, and a notification is displayed on the user's device screen.

[0380] Step 7:

[0381] The user provides feedback on the system's analysis results. The input is the user's feedback (e.g., "This analysis result is accurate"), and the output is the feedback data recorded on the server. Specifically, the user enters a comment using the feedback form and clicks the submit button. The server receives the feedback and records it in a log.

[0382] The above steps enable highly accurate fraud risk assessment and the generation of warning messages, allowing users to safely use social networking sites and dating applications.

[0383] (Application example 2)

[0384] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0385] In modern society, fraudulent activities via social networking sites and other information exchange platforms are on the rise, potentially causing financial losses to many people. Furthermore, fraudulent methods are becoming increasingly sophisticated, making it difficult for users to assess the risks involved in their own communications. Therefore, there is a need for a system that can detect these fraudulent activities early and warn users. The purpose of this invention is to solve these problems and ensure user safety.

[0386] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the user's emotions and adding an emotional evaluation to the warning message, means for notifying the user of the generated warning message, and means for generating a prompt sentence using a framework to evaluate the probability of fraud and the user's emotional state. This makes it possible to accurately evaluate the fraud risk and provide an appropriate warning that takes the user's emotional state into account in real time.

[0387] "User" refers to an entity (generally a human being) that uses a means of information exchange.

[0388] "Information exchange means" refers to systems that include software and hardware that allow users to communicate, such as social networking sites and communication apps.

[0389] "Central Computing Unit" means the central computer system that receives, analyzes, and stores data.

[0390] An "artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze input data and extract specific patterns or elements.

[0391] A "database" refers to an information management system that references and stores past data and specific patterns based on analysis results.

[0392] "Fraud" refers to any dishonest act that involves deceiving others to obtain money or information.

[0393] "Emotion analysis" refers to the technology of determining a user's emotional state from data such as text and voice.

[0394] A "prompt" is an instruction given to a generative AI model, a piece of text that provides context for a specific analysis or generation task.

[0395] An "encryption protocol" is a technology for ensuring the security of communications when transmitting data, encrypting the data to protect it from unauthorized access by third parties.

[0396] "Warning Message" means a message intended to inform users of fraud risks or other important information.

[0397] The system of the present invention analyzes past interactions of users of social networking sites and information exchange means, assesses fraud risk and emotional state, and issues warnings. This system is composed of a user terminal, a central computing device, an AI analysis system, a database, an emotion analysis engine, and a prompt sentence generation means.

[0398] System Overview

[0399] 1. User Device

[0400] Users can use their smartphones or computers to input past interactions on social media or communication apps. For example, they can copy conversations on a chat app and input them into the system.

[0401] 2. Central computing device

[0402] This is the central computer system that receives data and runs the AI ​​analysis system and sentiment analysis engine. It also manages the database and notifies results.

[0403] 3. Artificial Intelligence Analysis System

[0404] It uses machine learning algorithms to study past fraud cases and extract elements based on fraud patterns with high accuracy, identifying phrases and patterns that indicate fraudulent activity and assessing fraud risk.

[0405] 4. Sentiment Analysis Engine

[0406] Analyzes sentiment from user input data and classifies it into positive, negative, and neutral categories, providing additional information for fraud risk assessment.

[0407] 5. Database

[0408] It is an information collection system that stores patterns and past cases of fraudulent activity. The central computing device refers to this database to assess fraud risk.

[0409] 6. Prompt generation method

[0410] The framework generates prompts to assess the probability of fraud and the emotional state of the user. The generated prompts are then fed into an AI model for specific analysis and evaluation.

[0411] Program processing explanation

[0412] The central computing unit passes the data entered by the user to the AI ​​analysis system, which uses machine learning algorithms to analyze the data and assess the risk of fraud. At the same time, the emotion analysis engine analyzes the user's emotions and classifies them as positive, negative, or neutral. Based on the analysis results and the emotion assessment, the central computing unit generates a warning message and notifies the user's device. It also generates a prompt sentence including the emotion assessment and inputs it into the AI ​​model to improve the accuracy of the analysis.

[0413] Specific examples

[0414] Example 1: Social media interactions

[0415] A user copies a conversation on social media and inputs it into the system. The content is, "Hello! Would you like to meet sometime? Can you lend me a little money first?" The central computing device passes this text to the artificial intelligence analysis system, which extracts elements that indicate fraud. At the same time, the emotion analysis engine analyzes the user's emotional state and classifies it as negative. Finally, the central computing device sends a warning message to the user's device saying, "This person is likely to be a fraudster. Be careful not to lend them money."

[0416] Prompt Sentence Examples

[0417] "What is the fraud risk of this message? We use a fraud risk assessment model."

[0418] "What is the emotional state of this message? We use sentiment analysis models."

[0419] In this way, users can analyze past interactions, understand fraud risk and emotional state, and make risk assessments and take appropriate measures.

[0420] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0421] Step 1:

[0422] A user copies past interactions on social networking sites or other information exchange platforms and inputs them into their device. The input data is in text format. This text becomes input data that is later sent from the device to the central computing device for analysis.

[0423] Step 2:

[0424] The terminal transmits the input data to the central computing device, where it is encrypted using a secure communication protocol (e.g., SSL) to protect it from unauthorized access by third parties. The output is the encrypted text data received at the central computing device.

[0425] Step 3:

[0426] The central computing unit decodes the received data and passes it to the AI ​​analysis system. The input is the decoded text data, which is preprocessed for analysis. Preprocessing includes normalizing the data and removing unnecessary characters. The output is data converted into a form suitable for analysis.

[0427] Step 4:

[0428] An AI analysis system analyzes text data and extracts elements that indicate fraud. It uses a model trained on past fraud data to assess the fraud risk from the input data. The output is a fraud risk assessment result, specifically a fraud likelihood score and a list of related elements.

[0429] Step 5:

[0430] The central computing device passes the fraud risk assessment result to the emotion analysis engine, which analyzes the user's emotions. The input is the fraud risk assessment result and text data, which the emotion analysis engine analyzes to determine the emotional state (positive, negative, neutral). The output is the emotion assessment result.

[0431] Step 6:

[0432] The central computing device integrates the fraud risk assessment results and the emotion assessment results to generate a warning message. The input is the fraud risk score and the emotion state information, and based on these, it generates an appropriate warning message for the user. The output is the warning message.

[0433] Step 7:

[0434] The generated warning message is sent to the user terminal. The central computing device sends the warning message to the user terminal, and the user receives the warning message so that the user can take appropriate action regarding the fraud risk. The output is the warning message received by the user.

[0435] Step 8:

[0436] The central computing device uses the framework to generate prompt sentences for assessing the probability of fraud and the emotional state. The inputs are the fraud risk assessment results and the emotional assessment results, and specific prompt sentences are generated based on these. The output is the prompt sentence.

[0437] Step 9:

[0438] The generated prompt sentences are input into an artificial intelligence model to improve analysis accuracy. Based on the prompt sentences, additional analysis or model retraining is performed as needed. The output is an improved analysis model or new analysis results.

[0439] In this way, at each step, the terminal, central computing device, artificial intelligence analysis system, and emotion analysis engine work together to assess fraud risk and emotional state and provide appropriate warnings.

[0440] 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.

[0441] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0442] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0443] [Second embodiment]

[0444] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0445] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0446] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0447] 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.

[0448] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0449] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0450] 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.

[0451] 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.

[0452] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0453] 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.

[0454] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0455] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0456] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, and evaluates the possibility of fraud. Specific embodiments are described below.

[0457] System Overview

[0458] The system mainly consists of the following components:

[0459] 1. User Device

[0460] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[0461] 2. Server

[0462] The server is the central computer system that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[0463] 3. AI Analysis System

[0464] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[0465] 4. Database

[0466] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[0467] Program processing explanation

[0468] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0469] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[0470] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[0471] Based on the analysis results, the server extracts fraudster patterns from the database and assesses the fraud risk, for example, calculating that the probability of fraud is 80%.

[0472] The server generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check for more information." and is sent to the user's device.

[0473] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model.

[0474] Specific examples

[0475] Example 1: Social media interactions

[0476] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0477] 2. The user terminal sends this text to the server.

[0478] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0479] 4. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[0480] 5. The server creates a warning message saying, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[0481] Example 2: Interactions on a dating app

[0482] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[0483] 2. The user terminal sends this text to the server.

[0484] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0485] 4. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[0486] 5. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[0487] In this way, users can analyze the true meaning of interactions on social media and dating apps and prevent themselves from falling victim to fraud.

[0488] The processing flow will be explained below.

[0489] Step 1:

[0490] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0491] Step 2:

[0492] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[0493] Step 3:

[0494] The server receives the data sent from the user terminal and confirms that the data has been received.

[0495] Step 4:

[0496] Before the server can pass the received data to the AI ​​analysis system, it cleanses the text data and formats it in a way that makes it easier to analyze, for example by removing unnecessary spaces and special characters.

[0497] Step 5:

[0498] An AI analysis system analyzes incoming data and extracts elements based on scammer patterns, specifically identifying phrases that indicate fraudulent activity, such as "please lend me some money."

[0499] Step 6:

[0500] Based on the results of the AI ​​analysis system, the server extracts fraudster patterns from a database that records past fraud cases and typical patterns of fraudulent behavior.

[0501] Step 7:

[0502] The server will evaluate multiple factors (e.g., phrases used, frequency of interactions, inconsistencies, etc.) and calculate a fraud risk as a percentage. For example, it may determine that the fraud risk is 80%.

[0503] Step 8:

[0504] The server generates a warning message based on the fraud risk assessment result, such as "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[0505] Step 9:

[0506] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[0507] Step 10:

[0508] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[0509] Step 11:

[0510] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[0511] Step 12:

[0512] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model in the future.

[0513] This is the specific processing flow of this system, which allows users to detect in advance the risk of becoming a victim of fraud through interactions on social media or dating apps, and respond appropriately.

[0514] Example 1

[0515] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0516] In modern society, fraudulent activities involving the misuse of communication applications are becoming more common. Users of social networking sites and dating apps are particularly susceptible to the sophisticated tactics of fraudsters. In addition to standard security measures, there is a demand for systems that can accurately detect potential fraudulent activity and quickly warn users. However, conventional systems have difficulty efficiently detecting signs of fraud and incorporating user feedback to improve accuracy. The purpose of this invention is to solve these problems.

[0517] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0518] In this invention, the server includes: a means for a user to input past interactions in a communication application; a means for transmitting the input data to an information processing device; a means for the information processing device to input the data received from the user to a machine learning model; a means for the machine learning model to analyze the data and extract elements based on fraudulent activity patterns; a means for the information processing device to extract fraudulent activity patterns from a database based on the analysis results; a means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; and a means for notifying the user of the generated warning message. This enables the system to detect possible fraudulent activity with high accuracy and quickly issue a warning to the user. Furthermore, by adding a means for updating the machine learning model based on user feedback, the system can continuously improve its accuracy. As a result, users can use communication applications with peace of mind.

[0519] "User" refers to an individual or corporation that uses this system.

[0520] "Communication applications" refers to all software that allows messages and information to be exchanged over the Internet, such as social networking sites and dating apps.

[0521] "Past interactions" refers to the history of messages and chats that a user has had on a communication application.

[0522] "Information processing device" refers to a general term for a computer system that receives, analyzes, stores, and notifies data.

[0523] A "machine learning model" refers to an algorithm that uses artificial intelligence to learn patterns from past data and analyze and predict new data.

[0524] "Fraud patterns" refer to common behavioral characteristics or phrases extracted from past instances of fraud or malicious behavior.

[0525] "Analysis results" refer to the indicators and elements of fraudulent activity extracted by the machine learning model from the input data.

[0526] A "database" refers to a system that stores information necessary for analysis, such as patterns of fraudulent activity and past cases.

[0527] "Fraud risk" refers to the possibility of fraud or malicious intent calculated based on the analysis results.

[0528] The "warning message" refers to information sent to the user to call their attention based on the analysis results.

[0529] "Feedback" refers to the opinions and evaluations provided by users regarding the system's analysis results.

[0530] MODE FOR CARRYING OUT THE INVENTION

[0531] The present invention is a system that allows users to input their past interactions on social media or dating apps, analyzes that data with an AI analysis system, evaluates the possibility of fraud, and issues a warning to the user. The program processing of this system is explained in detail below.

[0532] System configuration

[0533] This invention mainly consists of the following hardware and software:

[0534] 1. User Device

[0535] A device that allows a user to input past interactions with a communication application, including a smartphone or computer.

[0536] 2. Server

[0537] The server is the central information processing device that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[0538] 3. AI Analysis System

[0539] The AI ​​analysis system analyzes data based on fraud patterns and extracts elements of fraudulent activity. It uses machine learning algorithms to enable highly accurate data analysis. For example, it uses machine learning frameworks such as TensorFlow and PyTorch.

[0540] 4. Database

[0541] It is an information collection system that stores fraud patterns and past cases. The server refers to this database to assess fraud risk.

[0542] Program processing overview

[0543] First, the user inputs a past conversation in a communication application in text format. For example, the user might input something like, "Hello! Would you like to meet sometime? Can you lend me a little money first?" into the system.

[0544] The user terminal then sends the entered data to the server, using an encryption protocol such as SSL to ensure the security of the communication.

[0545] The server passes the received data to an AI analysis system, which extracts elements that indicate fraudulent activity. The AI ​​analysis system learns from past cases of fraud and identifies and analyzes phrases such as "lend me some money."

[0546] Based on the analysis results, the server extracts fraudulent patterns from the database and assesses the fraud risk. For example, if the probability of fraud is calculated to be 80%, it generates a specific warning message to inform the user of the risk.

[0547] The generated warning message may say something like, "This person is likely to be a fraud. Avoid lending money to them and check for more information," and is sent from the server to the user's terminal.

[0548] Users can provide feedback on the analysis results. This feedback is sent to the server and used to improve the AI ​​model. For example, the feedback could be, "This warning was appropriate."

[0549] Specific examples

[0550] Example 1: Social media interactions

[0551] 1. The user enters the chat history from a social networking site into the system: "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[0552] 2. The user terminal sends this text to the server.

[0553] 3. The server passes the received data to the AI ​​analysis system.

[0554] 4. The server's AI analysis system determines that the phrase "lend me money" is 80% likely to be a scam.

[0555] 5. Based on the results, the server generates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and notifies the user.

[0556] 6. The user sends feedback saying, "This warning was appropriate."

[0557] 7. The server stores the feedback in a database and uses it for future analysis.

[0558] Example 2: Interactions on a dating app

[0559] 1. The user enters the chat history from the dating app into the system: "I want to get married soon! There's an investment opportunity. Are you interested?"

[0560] 2. The user device sends this data to the server.

[0561] 3. The server passes the received data to the AI ​​analysis system.

[0562] 4. The server's AI analysis system detects phrases related to "investment opportunities" and rates the likelihood of fraud at 90%.

[0563] 5. Based on the high fraud risk, the server generates a warning message stating, "This person is very likely to be a fraud. Please be especially careful with investment proposals and check the detailed information." and notifies the user.

[0564] 6. The user sends feedback saying, "This warning was helpful."

[0565] 7. The server stores the feedback in a database and uses it to improve the AI ​​model.

[0566] In this way, the system of the present invention can analyze the true intent of users' interactions in communication applications and prevent fraud before it occurs. Furthermore, by incorporating user feedback, the system's analysis accuracy can be continuously improved.

[0567] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0568] Step 1:

[0569] The user inputs past conversations in a communication application in text format. For example, the user copies and pastes the chat content from an SNS such as "Hello! Let's meet sometime. Can you lend me a little money first?" into the system. The input text is temporarily stored on the user's device.

[0570] Step 2:

[0571] The user terminal sends the entered text data to the server. At this time, security is ensured when sending data using encryption protocols such as SSL. Specifically, when the user presses the "Send" button, the text data is encrypted and sent to the server via the Internet. The input is the text data entered by the user, and the output is the encrypted data received by the server.

[0572] Step 3:

[0573] The server receives the data sent from the user terminal. The received data is decrypted and restored to the original text data. Specifically, the decryption program on the server decrypts the encrypted data according to the SSL protocol to obtain plain text data. The input is encrypted text data, and the output is decrypted text data.

[0574] Step 4:

[0575] The server passes the decrypted text data to the AI ​​analysis system. The passed data is added to a processing queue within the AI ​​analysis system. Specifically, the server converts the text data into a format suitable for the machine learning model and sends it to the analysis system. The input is the decrypted text data, and the output is the data input to the machine learning model.

[0576] Step 5:

[0577] The AI ​​analysis system analyzes the data and extracts elements that indicate fraudulent activity. For example, it identifies the phrase "lend me some money" and extracts corresponding patterns. The input is the text data to be analyzed, and the output is the extracted elements of fraudulent activity.

[0578] Step 6:

[0579] The server then references the database based on the analysis results and compares them with past fraudulent activity patterns. Specifically, the server executes a database query to search for relevant past cases. The input is the extracted fraudulent activity elements, and the output is the corresponding fraudulent activity patterns retrieved from the database.

[0580] Step 7:

[0581] The server assesses the fraud risk based on the fraudulent activity patterns and analysis results. This assessment is converted into a numerical value according to an algorithm, for example, the probability of fraud is calculated as 80%. Specifically, the server executes the risk assessment algorithm and calculates the fraud risk score. The input is the fraudulent activity patterns and analysis results obtained from the database, and the output is the fraud risk score.

[0582] Step 8:

[0583] The server generates a warning message based on the fraud risk. For example, it creates a warning message that reads, "This person is likely to be a fraudster. Avoid lending money to them and check for more information." The input is the fraud risk score, and the output is the generated warning message.

[0584] Step 9:

[0585] The server sends the generated warning message to the user terminal. The sent message is displayed on the user terminal. Specifically, the server sends a message to the user terminal using a message transmission protocol, and the user terminal displays the message on its screen. The input is the generated warning message, and the output is the warning message displayed on the user terminal.

[0586] Step 10:

[0587] The user provides feedback on the analysis results. This feedback is sent from the user terminal to the server. For example, the user may input "This warning was appropriate." The input is the feedback on the analysis results, and the output is the feedback data sent to the server.

[0588] Step 11:

[0589] The server receives user feedback and updates the machine learning model of the AI ​​analysis system. Specifically, the server passes the feedback data to the AI ​​analysis system and retrains the machine learning algorithm. The input is the feedback data, and the output is the updated machine learning model.

[0590] (Application example 1)

[0591] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0592] Users of communication platforms such as social networking sites and dating apps are at increased risk of becoming victims of fraud. Under these circumstances, there is a need for a method that automatically and accurately analyzes users' risk of fraud based on their past interactions and prevents fraud from occurring. However, with conventional methods, it is difficult for users to determine the risk themselves, which often results in delayed response. The present invention aims to solve this problem.

[0593] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0594] In this invention, the server includes: means for a user to input past interactions on the communication platform; means for transmitting the input data; means for the information processing device to input the data received from the user to a knowledge base analysis system; means for the knowledge base analysis system to analyze the data and extract elements based on fraud patterns; means for the information processing device to extract fraud patterns from the information accumulation system based on the analysis results; means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; means for notifying the user of the generated warning message; a mobile terminal having a user interface for copying and pasting interactions into an application; and a data transmission module having a protocol for encrypting and transmitting the input data. This enables a user to easily input past interactions and have their risks quickly and accurately evaluated.

[0595] "User" refers to any individual or organization using the communication platform.

[0596] A "communication platform" refers to an online service that allows users to exchange messages with each other, such as social networking sites or matching apps.

[0597] "Interaction" refers to the sending and receiving of messages between users on the communication platform.

[0598] "Input means" refers to a method or device that allows a user to provide the content of an interaction to the system in text form.

[0599] "Information processing device" refers to a computer device that receives, processes, and transmits data.

[0600] "Knowledge-based analysis system" refers to an artificial intelligence system for analyzing data based on patterns of fraudulent activity.

[0601] "Means for extracting elements" refers to the methods and devices that the knowledge-based analysis system uses to find features and patterns related to fraudulent activity in the data.

[0602] An "information collection system" refers to a database that accumulates and manages fraud patterns and past cases.

[0603] "Warning message" refers to a message intended to inform users of the risk of fraud.

[0604] "User interface" refers to the screens and input devices through which a user interacts with a system and inputs data.

[0605] "Mobile terminal" refers to an electronic device that a user can carry with them, such as a smartphone or tablet.

[0606] An "encryption and transmission protocol" refers to a method or communication protocol for encrypting and transmitting data to ensure secure data communication.

[0607] The system of the present invention allows users to input their past interactions on a communication platform, analyzes their hidden intentions using a knowledge-based analysis system, and assesses the risk of fraud. Specific embodiments are described below.

[0608] System Overview

[0609] The system mainly consists of the following components:

[0610] 1. User Device

[0611] A mobile device for inputting past interactions on a communication platform via a user interface. Users use a smartphone or tablet to input the relevant text into the system.

[0612] 2. Information Processing Device

[0613] It is a central computer system that receives data and executes the knowledge-based analysis system. The information processing device also manages the information collection system and notifies the results.

[0614] 3. Knowledge-based analysis system

[0615] The system analyzes data based on fraud patterns and extracts relevant elements. The knowledge-based analysis system uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[0616] 4. Information Collection System

[0617] This is a database for storing fraud patterns and past cases. The information processing device refers to this information accumulation system to assess fraud risk.

[0618] Processing Step Description

[0619] First, the user inputs a past conversation on the communication platform in text format, for example, "Hello! Would you like to meet up sometime? Can you lend me a little money first?" and pastes it into the application.

[0620] Next, the user terminal transmits the input data to the information processing device. This data transmission uses an encryption protocol such as SSL to ensure the security of the communication.

[0621] The information processing device passes the received data to a knowledge-based analysis system, which extracts elements that indicate fraud. The knowledge-based analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[0622] Based on the analysis results, the information processing device extracts fraud patterns from the information collection system and evaluates the fraud risk. For example, it calculates that the probability of fraud is 80%.

[0623] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and is sent to the user terminal.

[0624] Additionally, users can provide feedback on the analysis results, which is sent to the information processing device and used to improve the knowledge-based analysis system.

[0625] Specific examples

[0626] Example 1: Analysis of social media interactions

[0627] Type: "Hi! Want to meet up sometime? Can I borrow a little money first?"

[0628] Analysis result: 80% fraud risk

[0629] Notification message: "This person is likely a scammer. Avoid lending money and check for more information."

[0630] Example 2: Analysis of interactions on a dating app

[0631] Type: "I'd love to get married soon! I have an investment opportunity, would you be interested?"

[0632] Analysis result: 90% fraud risk

[0633] Notification message: "This person is very likely a scam. Be especially careful with investment proposals and check the full details."

[0634] Prompt Sentence Examples

[0635] "Analyze the following exchange and assess the risk of fraud: 'Hi! Would you like to meet up sometime? Can I borrow a little money first?'"

[0636] Analyze the following exchange and assess its risk of fraud: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[0637] In this way, users can analyze the true meaning of interactions on the communication platform and prevent themselves from falling victim to fraud.

[0638] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0639] Step 1:

[0640] A user inputs past interactions on a communication platform in text format. The user inputs (copy and paste) the text via the user interface of a smartphone or tablet, and prepares this text for subsequent analysis. The input includes content such as "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[0641] Step 2:

[0642] The user terminal transmits the input data to the information processing device. An encryption protocol such as SSL is used for data transmission to ensure communication security. The input in this step is the text entered by the user, and the output is the transmission of encrypted data.

[0643] Step 3:

[0644] The information processing device passes the received data to a knowledge-based analysis system. This data is decrypted in its original text form and provided to the knowledge-based analysis system for analysis. The input is the decrypted content of the encrypted data, and the output is analyzable text data.

[0645] Step 4:

[0646] A knowledge-based analysis system analyzes the data and extracts indicators of fraud. It uses a generative AI model trained on past fraud cases to identify phrases such as "lend me some money." The input is analyzable text data, and the output is a fraud risk score and indicators of fraud.

[0647] Step 5:

[0648] The data processor extracts fraud patterns from the data collection system based on the analysis results and assesses the fraud risk. The data collection system stores past fraud patterns and cases, and compares them with the current input data. The input is a fraud risk score and fraud indicator elements, and the output is a final fraud risk assessment.

[0649] Step 6:

[0650] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. The generated warning message includes content such as "This person is likely to be a fraudster. Avoid lending money and check the detailed information." The input is the fraud risk assessment, and the output is a warning message to the user.

[0651] Step 7:

[0652] The user provides feedback on the analysis results. The feedback is sent back to the information processing system, and the knowledge-based analysis system updates the model based on this feedback. The input in this step is the user's feedback, and the output is the updated analytical model.

[0653] This process allows users to quickly and automatically analyze whether their interactions on the communication platform pose a risk of fraud and receive appropriate warnings.

[0654] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0655] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, evaluates the possibility of fraud, and provides more accurate warnings by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments are described below.

[0656] System Overview

[0657] The system mainly consists of the following components:

[0658] 1. User Device

[0659] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[0660] 2. Server

[0661] It is the central computer system that receives data and runs the AI ​​analysis system and emotion engine. The server also manages the database and notifies the results.

[0662] 3. AI Analysis System

[0663] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[0664] 4. Emotion Engine

[0665] Analyzing emotions from user input data and classifying them into positive, negative, and neutral categories, the emotion engine provides emotional information to the AI ​​analysis system, which then adds it to the fraud risk assessment.

[0666] 5. Database

[0667] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[0668] Program processing explanation

[0669] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0670] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[0671] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[0672] At the same time, the server also passes the received data to the emotion engine, which analyzes the user's emotions. The emotion engine analyzes the emotions in the text and classifies them as positive, negative, or neutral.

[0673] Based on the analysis results, the server extracts fraudster patterns from the database and evaluates the fraud risk. Emotional information from the emotion engine is also added to the fraud risk assessment. For example, the probability of fraud is 80% and the user's emotion is determined to be negative.

[0674] The server generates a warning message based on the fraud risk assessment results and emotional information and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and sent to the user's terminal.

[0675] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model and emotion engine.

[0676] Specific examples

[0677] Example 1: Social media interactions

[0678] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0679] 2. The user terminal sends this text to the server.

[0680] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0681] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as negative.

[0682] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[0683] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[0684] Example 2: Interactions on a dating app

[0685] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[0686] 2. The user terminal sends this text to the server.

[0687] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0688] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as neutral.

[0689] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[0690] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[0691] In this way, users can analyze their interactions on social media and dating apps to understand their emotional state as well as the risk of fraud, enabling more accurate risk assessment and appropriate responses.

[0692] The processing flow will be explained below.

[0693] Step 1:

[0694] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0695] Step 2:

[0696] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[0697] Step 3:

[0698] The server receives the data sent from the user terminal and confirms that the data has been received.

[0699] Step 4:

[0700] The server cleanses the incoming data and formats it in a way that makes it easier to parse, for example by removing extra whitespace and special characters.

[0701] Step 5:

[0702] The server passes the cleansed data to the AI ​​analysis system, which then analyzes the data and extracts elements that indicate fraudulent activity, such as requests to lend money.

[0703] Step 6:

[0704] The server then passes the same cleansed data to a sentiment engine, which analyzes the data and identifies sentiments such as positive, negative, or neutral.

[0705] Step 7:

[0706] The emotion engine returns the analysis results to the server and determines whether the user has a negative emotion. For example, the emotion engine classifies the emotion as "negative."

[0707] Step 8:

[0708] The server extracts fraudster patterns from the database based on the results of the AI ​​analysis system and the emotion engine, and matches patterns that contain elements that indicate fraudulent activity.

[0709] Step 9:

[0710] The server comprehensively evaluates the fraud risk based on the analysis results and emotional information. For example, if the fraud rate is 80% and the user's emotional state is negative,

[0711] Step 10:

[0712] The server generates a warning message based on the fraud risk assessment results and emotional information. For example, it creates a message saying, "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[0713] Step 11:

[0714] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[0715] Step 12:

[0716] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[0717] Step 13:

[0718] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[0719] Step 14:

[0720] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model and emotion engine in the future.

[0721] This is the specific processing flow of this system. As a result, users can detect fraudulent behavior early through their interactions on social media or dating apps, as well as recognize their own emotional state and take appropriate action.

[0722] Example 2

[0723] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0724] In conventional social networking services and dating applications, there was no system that effectively assessed and warned users about their risk of being scammed. As a result, users were more likely to fall prey to scammers' sophisticated tactics, increasing the risk of financial loss. In addition, the accuracy of warnings was low because the system did not take into account the user's emotional state.

[0725] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to input past interactions on a social networking service or a matching application; a means for transmitting the input data; a means for the processing device to input data received from the user to an AI analysis system; a means for the AI ​​analysis system to analyze the data and extract elements based on fraud patterns; a means for the processing device to extract fraud patterns from the data management system based on the analysis results; a means for the processing device to evaluate fraud risk based on the analysis results and generate a warning message; a means for notifying the user of the generated warning message; and a means for analyzing emotions in the input data and adding emotion-based information to the fraud risk evaluation. This allows users to quickly and accurately evaluate the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[0726] "User" refers to an individual who uses a social networking service or dating application.

[0727] "Social networking service" refers to an online platform that enables users to communicate with other individuals.

[0728] "Matching Application" means a mobile or web application that enables users to meet and connect with other individuals.

[0729] "Past interactions" refers to the history of messages exchanged by a user on social networking services or matching applications.

[0730] "Means for input" refers to an interface or device that allows a user to send past interactions to the system in text form.

[0731] "Processing device" refers to a central computer system for receiving, analyzing, and evaluating input data.

[0732] "Artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze data and detect patterns of fraudulent activity.

[0733] A "data management system" refers to an information storage system that stores past fraud cases and patterns and references them as needed.

[0734] "Fraud" refers to any dishonest activity that attempts to deceive users into giving them money or personal information.

[0735] "Elements" refer to the characteristics extracted by artificial intelligence analysis systems to identify patterns of fraudulent activity.

[0736] "Analysis results" refers to the conclusions and data derived by the AI ​​analysis system through data analysis.

[0737] "Fraud risk" refers to the degree of likelihood that the input data represents fraudulent activity.

[0738] "Warning Message" refers to a message containing a warning provided to a user based on a fraud risk assessment.

[0739] "Means of notification" refers to the communication means or interface used to convey a warning message to the user.

[0740] "Means for analyzing emotions" refers to systems or algorithms for extracting emotional information from text data and analyzing that information.

[0741] "Emotion-based information" refers to data that represents a user's emotional state and is used in fraud risk assessment.

[0742] This invention provides a system that inputs past interactions a user has had on social networking services or matching applications, analyzes the data using an AI analysis system and a sentiment analysis engine, and assesses the risk of fraud. This system is comprised of a user terminal, a server, an AI analysis system, a sentiment analysis engine, and a database.

[0743] Hardware and software used

[0744] User device: The smartphone or computer that the user uses to enter data and receive results. Examples: iPhone, Android device, Windows PC, Mac.

[0745] Server: A high-performance computer system to receive data and perform analytical processing. Examples: AWS EC2, Google Cloud Compute Engine.

[0746] Communication Protocol: The secure communication method used to send and receive data. For example: SSL / TLS.

[0747] Artificial intelligence analysis systems: Use machine learning algorithms that learn fraud patterns. Examples: Scikit-learn, TensorFlow.

[0748] Sentiment analysis engine: uses algorithms that analyze user sentiment. Example: IBM Watson Natural Language Understanding.

[0749] Database: A system for storing information on past fraud cases and patterns. Examples: MySQL, PostgreSQL.

[0750] System Overview

[0751] 1. User enters data:

[0752] The user inputs past interactions in text format, for example, "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[0753] 2. Data transmission:

[0754] The user terminal sends the entered data to the server using an encryption protocol (such as SSL / TLS).

[0755] 3. Data Receipt and Analysis:

[0756] The server receives the data and first analyzes it with an artificial intelligence analysis system, which uses machine learning algorithms that have learned fraudulent patterns to extract elements, such as phrases like "lend me some money."

[0757] 4. Emotion analysis:

[0758] The data is passed to a sentiment analysis engine to analyze the user's emotional state. The text is classified as positive, negative, or neutral. For example, if the user's message is offensive, it will be classified as negative.

[0759] 5. Fraud risk assessment and warning message generation:

[0760] The server integrates the results of the AI ​​analysis system and the sentiment analysis engine, and compares them with past fraud patterns in the database to assess the fraud risk. For example, if the fraud risk is determined to be high (80%), a warning message is generated based on that risk.

[0761] 6. Warning message notification:

[0762] The server generates a warning message and sends it to the user's device, which displays the message "This person is likely a fraud. Avoid lending money to them and check for more information."

[0763] 7. Gathering Feedback:

[0764] Users provide feedback on the analysis results, which is sent to the server and used to improve the artificial intelligence model and sentiment analysis engine.

[0765] Specific examples

[0766] Example 1: Social media interactions

[0767] 1. Users copy social media conversations and enter them into the system.

[0768] For example: "Hi! Would you like to meet up sometime? Can I borrow a little money first?"

[0769] 2. The user terminal sends this text to the server.

[0770] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[0771] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as negative.

[0772] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[0773] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[0774] Example 2: Interactions on a dating app

[0775] 1. The user copies the conversations they have had on the dating app and enters them into the system.

[0776] Example: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[0777] 2. The user terminal sends this text to the server.

[0778] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[0779] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as neutral.

[0780] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[0781] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[0782] This allows users to quickly and accurately assess the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[0783] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0784] Step 1:

[0785] The user inputs past interactions in text format on social networking services or dating applications. Specifically, the user inputs past messages by copying and pasting them into the application screen on their smartphone or computer. An example of input is the message "Hello! Would you like to meet sometime? Can you lend me a little money first?" The input data is saved in text format.

[0786] Step 2:

[0787] The user terminal sends the entered text data to the server. At this time, an encryption protocol such as SSL / TLS is used to ensure the security of the communication. The input is text data, and the output is encrypted data. In concrete terms, the application on the user terminal issues a send request to the server, and a progress bar is displayed.

[0788] Step 3:

[0789] The server passes the received data to an artificial intelligence analysis system, which extracts elements that indicate fraud. The data input is encrypted text data, and the output contains elements related to fraud patterns (e.g., keywords such as "money" and "lend me"). This processing uses machine learning algorithms (e.g., Scikit-learn, TensorFlow). Specifically, the server records a "receiving data" log and displays an analysis progress bar in percentage.

[0790] Step 4:

[0791] At the same time, the server passes the received data to a sentiment analysis engine to analyze the user's sentiment. The data input is the original text data, and the output is the sentiment classification result (positive, negative, neutral). A sentiment analysis algorithm (e.g., IBM Watson Natural Language Understanding) is used for sentiment analysis. Specifically, the server records a log of "sentiment analysis in progress" and displays the progress of sentiment analysis in percentage.

[0792] Step 5:

[0793] The server integrates the results of the AI ​​analysis system and sentiment analysis, and compares them with past fraud patterns in the database to assess the fraud risk. The input is the elements of the fraud pattern and the sentiment classification results, and the output is the fraud risk assessment result (e.g., fraud risk 80%). The server generates a warning message based on the analysis results. Specifically, the server records a log of "fraud risk assessment in progress" and displays the fraud risk value.

[0794] Step 6:

[0795] The server generates a warning message and sends it to the user's device. The input is the fraud risk assessment result, and the output is a warning message (e.g., "This person is likely to be a fraudster. Avoid lending money and check for more information."). Specifically, the server issues a message sending request to the user's device, and a notification is displayed on the user's device screen.

[0796] Step 7:

[0797] The user provides feedback on the system's analysis results. The input is the user's feedback (e.g., "This analysis result is accurate"), and the output is the feedback data recorded on the server. Specifically, the user enters a comment using the feedback form and clicks the submit button. The server receives the feedback and records it in a log.

[0798] The above steps enable highly accurate fraud risk assessment and the generation of warning messages, allowing users to safely use social networking sites and dating applications.

[0799] (Application example 2)

[0800] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0801] In modern society, fraudulent activities via social networking sites and other information exchange platforms are on the rise, potentially causing financial losses to many people. Furthermore, fraudulent methods are becoming increasingly sophisticated, making it difficult for users to assess the risks involved in their own communications. Therefore, there is a need for a system that can detect these fraudulent activities early and warn users. The purpose of this invention is to solve these problems and ensure user safety.

[0802] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the user's emotions and adding an emotional evaluation to the warning message, means for notifying the user of the generated warning message, and means for generating a prompt sentence using a framework to evaluate the probability of fraud and the user's emotional state. This makes it possible to accurately evaluate the fraud risk and provide an appropriate warning that takes the user's emotional state into account in real time.

[0803] "User" refers to an entity (generally a human being) that uses a means of information exchange.

[0804] "Information exchange means" refers to systems that include software and hardware that allow users to communicate, such as social networking sites and communication apps.

[0805] "Central Computing Unit" means the central computer system that receives, analyzes, and stores data.

[0806] An "artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze input data and extract specific patterns or elements.

[0807] A "database" refers to an information management system that references and stores past data and specific patterns based on analysis results.

[0808] "Fraud" refers to any dishonest act that involves deceiving others to obtain money or information.

[0809] "Emotion analysis" refers to the technology of determining a user's emotional state from data such as text and voice.

[0810] A "prompt" is an instruction given to a generative AI model, a piece of text that provides context for a specific analysis or generation task.

[0811] An "encryption protocol" is a technology for ensuring the security of communications when transmitting data, encrypting the data to protect it from unauthorized access by third parties.

[0812] "Warning Message" means a message intended to inform users of fraud risks or other important information.

[0813] The system of the present invention analyzes past interactions of users of social networking sites and information exchange means, assesses fraud risk and emotional state, and issues warnings. This system is composed of a user terminal, a central computing device, an AI analysis system, a database, an emotion analysis engine, and a prompt sentence generation means.

[0814] System Overview

[0815] 1. User Device

[0816] Users can use their smartphones or computers to input past interactions on social media or communication apps. For example, they can copy conversations on a chat app and input them into the system.

[0817] 2. Central computing device

[0818] This is the central computer system that receives data and runs the AI ​​analysis system and sentiment analysis engine. It also manages the database and notifies results.

[0819] 3. Artificial Intelligence Analysis System

[0820] It uses machine learning algorithms to study past fraud cases and extract elements based on fraud patterns with high accuracy, identifying phrases and patterns that indicate fraudulent activity and assessing fraud risk.

[0821] 4. Sentiment Analysis Engine

[0822] Analyzes sentiment from user input data and classifies it into positive, negative, and neutral categories, providing additional information for fraud risk assessment.

[0823] 5. Database

[0824] It is an information collection system that stores patterns and past cases of fraudulent activity. The central computing device refers to this database to assess fraud risk.

[0825] 6. Prompt generation method

[0826] The framework generates prompts to assess the probability of fraud and the emotional state of the user. The generated prompts are then fed into an AI model for specific analysis and evaluation.

[0827] Program processing explanation

[0828] The central computing unit passes the data entered by the user to the AI ​​analysis system, which uses machine learning algorithms to analyze the data and assess the risk of fraud. At the same time, the emotion analysis engine analyzes the user's emotions and classifies them as positive, negative, or neutral. Based on the analysis results and the emotion assessment, the central computing unit generates a warning message and notifies the user's device. It also generates a prompt sentence including the emotion assessment and inputs it into the AI ​​model to improve the accuracy of the analysis.

[0829] Specific examples

[0830] Example 1: Social media interactions

[0831] A user copies a conversation on social media and inputs it into the system. The content is, "Hello! Would you like to meet sometime? Can you lend me a little money first?" The central computing device passes this text to the artificial intelligence analysis system, which extracts elements that indicate fraud. At the same time, the emotion analysis engine analyzes the user's emotional state and classifies it as negative. Finally, the central computing device sends a warning message to the user's device saying, "This person is likely to be a fraudster. Be careful not to lend them money."

[0832] Prompt Sentence Examples

[0833] "What is the fraud risk of this message? We use a fraud risk assessment model."

[0834] "What is the emotional state of this message? We use sentiment analysis models."

[0835] In this way, users can analyze past interactions, understand fraud risk and emotional state, and make risk assessments and take appropriate measures.

[0836] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0837] Step 1:

[0838] A user copies past interactions on social networking sites or other information exchange platforms and inputs them into their device. The input data is in text format. This text becomes input data that is later sent from the device to the central computing device for analysis.

[0839] Step 2:

[0840] The terminal transmits the input data to the central computing device, where it is encrypted using a secure communication protocol (e.g., SSL) to protect it from unauthorized access by third parties. The output is the encrypted text data received at the central computing device.

[0841] Step 3:

[0842] The central computing unit decodes the received data and passes it to the AI ​​analysis system. The input is the decoded text data, which is preprocessed for analysis. Preprocessing includes normalizing the data and removing unnecessary characters. The output is data converted into a form suitable for analysis.

[0843] Step 4:

[0844] An AI analysis system analyzes text data and extracts elements that indicate fraud. It uses a model trained on past fraud data to assess the fraud risk from the input data. The output is a fraud risk assessment result, specifically a fraud likelihood score and a list of related elements.

[0845] Step 5:

[0846] The central computing device passes the fraud risk assessment result to the emotion analysis engine, which analyzes the user's emotions. The input is the fraud risk assessment result and text data, which the emotion analysis engine analyzes to determine the emotional state (positive, negative, neutral). The output is the emotion assessment result.

[0847] Step 6:

[0848] The central computing device integrates the fraud risk assessment results and the emotion assessment results to generate a warning message. The input is the fraud risk score and the emotion state information, and based on these, it generates an appropriate warning message for the user. The output is the warning message.

[0849] Step 7:

[0850] The generated warning message is sent to the user terminal. The central computing device sends the warning message to the user terminal, and the user receives the warning message so that the user can take appropriate action regarding the fraud risk. The output is the warning message received by the user.

[0851] Step 8:

[0852] The central computing device uses the framework to generate prompt sentences for assessing the probability of fraud and the emotional state. The inputs are the fraud risk assessment results and the emotional assessment results, and specific prompt sentences are generated based on these. The output is the prompt sentence.

[0853] Step 9:

[0854] The generated prompt sentences are input into an artificial intelligence model to improve analysis accuracy. Based on the prompt sentences, additional analysis or model retraining is performed as needed. The output is an improved analysis model or new analysis results.

[0855] In this way, at each step, the terminal, central computing device, artificial intelligence analysis system, and emotion analysis engine work together to assess fraud risk and emotional state and provide appropriate warnings.

[0856] 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.

[0857] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0858] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0859] [Third embodiment]

[0860] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0861] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0862] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0863] 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.

[0864] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0865] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0866] 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.

[0867] 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.

[0868] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0869] 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.

[0870] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0871] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0872] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, and evaluates the possibility of fraud. Specific embodiments are described below.

[0873] System Overview

[0874] The system mainly consists of the following components:

[0875] 1. User Device

[0876] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[0877] 2. Server

[0878] The server is the central computer system that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[0879] 3. AI Analysis System

[0880] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[0881] 4. Database

[0882] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[0883] Program processing explanation

[0884] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0885] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[0886] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[0887] Based on the analysis results, the server extracts fraudster patterns from the database and assesses the fraud risk, for example, calculating that the probability of fraud is 80%.

[0888] The server generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check for more information." and is sent to the user's device.

[0889] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model.

[0890] Specific examples

[0891] Example 1: Social media interactions

[0892] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0893] 2. The user terminal sends this text to the server.

[0894] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0895] 4. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[0896] 5. The server creates a warning message saying, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[0897] Example 2: Interactions on a dating app

[0898] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[0899] 2. The user terminal sends this text to the server.

[0900] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[0901] 4. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[0902] 5. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[0903] In this way, users can analyze the true meaning of interactions on social media and dating apps and prevent themselves from falling victim to fraud.

[0904] The processing flow will be explained below.

[0905] Step 1:

[0906] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[0907] Step 2:

[0908] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[0909] Step 3:

[0910] The server receives the data sent from the user terminal and confirms that the data has been received.

[0911] Step 4:

[0912] Before the server can pass the received data to the AI ​​analysis system, it cleanses the text data and formats it in a way that makes it easier to analyze, for example by removing unnecessary spaces and special characters.

[0913] Step 5:

[0914] An AI analysis system analyzes incoming data and extracts elements based on scammer patterns, specifically identifying phrases that indicate fraudulent activity, such as "please lend me some money."

[0915] Step 6:

[0916] Based on the results of the AI ​​analysis system, the server extracts fraudster patterns from a database that records past fraud cases and typical patterns of fraudulent behavior.

[0917] Step 7:

[0918] The server will evaluate multiple factors (e.g., phrases used, frequency of interactions, inconsistencies, etc.) and calculate a fraud risk as a percentage. For example, it may determine that the fraud risk is 80%.

[0919] Step 8:

[0920] The server generates a warning message based on the fraud risk assessment result, such as "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[0921] Step 9:

[0922] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[0923] Step 10:

[0924] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[0925] Step 11:

[0926] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[0927] Step 12:

[0928] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model in the future.

[0929] This is the specific processing flow of this system, which allows users to detect in advance the risk of becoming a victim of fraud through interactions on social media or dating apps, and respond appropriately.

[0930] Example 1

[0931] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0932] In modern society, fraudulent activities involving the misuse of communication applications are becoming more common. Users of social networking sites and dating apps are particularly susceptible to the sophisticated tactics of fraudsters. In addition to standard security measures, there is a demand for systems that can accurately detect potential fraudulent activity and quickly warn users. However, conventional systems have difficulty efficiently detecting signs of fraud and incorporating user feedback to improve accuracy. The purpose of this invention is to solve these problems.

[0933] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0934] In this invention, the server includes: a means for a user to input past interactions in a communication application; a means for transmitting the input data to an information processing device; a means for the information processing device to input the data received from the user to a machine learning model; a means for the machine learning model to analyze the data and extract elements based on fraudulent activity patterns; a means for the information processing device to extract fraudulent activity patterns from a database based on the analysis results; a means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; and a means for notifying the user of the generated warning message. This enables the system to detect possible fraudulent activity with high accuracy and quickly issue a warning to the user. Furthermore, by adding a means for updating the machine learning model based on user feedback, the system can continuously improve its accuracy. As a result, users can use communication applications with peace of mind.

[0935] "User" refers to an individual or corporation that uses this system.

[0936] "Communication applications" refers to all software that allows messages and information to be exchanged over the Internet, such as social networking sites and dating apps.

[0937] "Past interactions" refers to the history of messages and chats that a user has had on a communication application.

[0938] "Information processing device" refers to a general term for a computer system that receives, analyzes, stores, and notifies data.

[0939] A "machine learning model" refers to an algorithm that uses artificial intelligence to learn patterns from past data and analyze and predict new data.

[0940] "Fraud patterns" refer to common behavioral characteristics or phrases extracted from past instances of fraud or malicious behavior.

[0941] "Analysis results" refer to the indicators and elements of fraudulent activity extracted by the machine learning model from the input data.

[0942] A "database" refers to a system that stores information necessary for analysis, such as patterns of fraudulent activity and past cases.

[0943] "Fraud risk" refers to the possibility of fraud or malicious intent calculated based on the analysis results.

[0944] The "warning message" refers to information sent to the user to call their attention based on the analysis results.

[0945] "Feedback" refers to the opinions and evaluations provided by users regarding the system's analysis results.

[0946] MODE FOR CARRYING OUT THE INVENTION

[0947] The present invention is a system that allows users to input their past interactions on social media or dating apps, analyzes that data with an AI analysis system, evaluates the possibility of fraud, and issues a warning to the user. The program processing of this system is explained in detail below.

[0948] System configuration

[0949] This invention mainly consists of the following hardware and software:

[0950] 1. User Device

[0951] A device that allows a user to input past interactions with a communication application, including a smartphone or computer.

[0952] 2. Server

[0953] The server is the central information processing device that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[0954] 3. AI Analysis System

[0955] The AI ​​analysis system analyzes data based on fraud patterns and extracts elements of fraudulent activity. It uses machine learning algorithms to enable highly accurate data analysis. For example, it uses machine learning frameworks such as TensorFlow and PyTorch.

[0956] 4. Database

[0957] It is an information collection system that stores fraud patterns and past cases. The server refers to this database to assess fraud risk.

[0958] Program processing overview

[0959] First, the user inputs a past conversation in a communication application in text format. For example, the user might input something like, "Hello! Would you like to meet sometime? Can you lend me a little money first?" into the system.

[0960] The user terminal then sends the entered data to the server, using an encryption protocol such as SSL to ensure the security of the communication.

[0961] The server passes the received data to an AI analysis system, which extracts elements that indicate fraudulent activity. The AI ​​analysis system learns from past cases of fraud and identifies and analyzes phrases such as "lend me some money."

[0962] Based on the analysis results, the server extracts fraudulent patterns from the database and assesses the fraud risk. For example, if the probability of fraud is calculated to be 80%, it generates a specific warning message to inform the user of the risk.

[0963] The generated warning message may say something like, "This person is likely to be a fraud. Avoid lending money to them and check for more information," and is sent from the server to the user's terminal.

[0964] Users can provide feedback on the analysis results. This feedback is sent to the server and used to improve the AI ​​model. For example, the feedback could be, "This warning was appropriate."

[0965] Specific examples

[0966] Example 1: Social media interactions

[0967] 1. The user enters the chat history from a social networking site into the system: "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[0968] 2. The user terminal sends this text to the server.

[0969] 3. The server passes the received data to the AI ​​analysis system.

[0970] 4. The server's AI analysis system determines that the phrase "lend me money" is 80% likely to be a scam.

[0971] 5. Based on the results, the server generates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and notifies the user.

[0972] 6. The user sends feedback saying, "This warning was appropriate."

[0973] 7. The server stores the feedback in a database and uses it for future analysis.

[0974] Example 2: Interactions on a dating app

[0975] 1. The user enters the chat history from the dating app into the system: "I want to get married soon! There's an investment opportunity. Are you interested?"

[0976] 2. The user device sends this data to the server.

[0977] 3. The server passes the received data to the AI ​​analysis system.

[0978] 4. The server's AI analysis system detects phrases related to "investment opportunities" and rates the likelihood of fraud at 90%.

[0979] 5. Based on the high fraud risk, the server generates a warning message stating, "This person is very likely to be a fraud. Please be especially careful with investment proposals and check the detailed information." and notifies the user.

[0980] 6. The user sends feedback saying, "This warning was helpful."

[0981] 7. The server stores the feedback in a database and uses it to improve the AI ​​model.

[0982] In this way, the system of the present invention can analyze the true intent of users' interactions in communication applications and prevent fraud before it occurs. Furthermore, by incorporating user feedback, the system's analysis accuracy can be continuously improved.

[0983] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0984] Step 1:

[0985] The user inputs past conversations in a communication application in text format. For example, the user copies and pastes the chat content from an SNS such as "Hello! Let's meet sometime. Can you lend me a little money first?" into the system. The input text is temporarily stored on the user's device.

[0986] Step 2:

[0987] The user terminal sends the entered text data to the server. At this time, security is ensured when sending data using encryption protocols such as SSL. Specifically, when the user presses the "Send" button, the text data is encrypted and sent to the server via the Internet. The input is the text data entered by the user, and the output is the encrypted data received by the server.

[0988] Step 3:

[0989] The server receives the data sent from the user terminal. The received data is decrypted and restored to the original text data. Specifically, the decryption program on the server decrypts the encrypted data according to the SSL protocol to obtain plain text data. The input is encrypted text data, and the output is decrypted text data.

[0990] Step 4:

[0991] The server passes the decrypted text data to the AI ​​analysis system. The passed data is added to a processing queue within the AI ​​analysis system. Specifically, the server converts the text data into a format suitable for the machine learning model and sends it to the analysis system. The input is the decrypted text data, and the output is the data input to the machine learning model.

[0992] Step 5:

[0993] The AI ​​analysis system analyzes the data and extracts elements that indicate fraudulent activity. For example, it identifies the phrase "lend me some money" and extracts corresponding patterns. The input is the text data to be analyzed, and the output is the extracted elements of fraudulent activity.

[0994] Step 6:

[0995] The server then references the database based on the analysis results and compares them with past fraudulent activity patterns. Specifically, the server executes a database query to search for relevant past cases. The input is the extracted fraudulent activity elements, and the output is the corresponding fraudulent activity patterns retrieved from the database.

[0996] Step 7:

[0997] The server assesses the fraud risk based on the fraudulent activity patterns and analysis results. This assessment is converted into a numerical value according to an algorithm, for example, the probability of fraud is calculated as 80%. Specifically, the server executes the risk assessment algorithm and calculates the fraud risk score. The input is the fraudulent activity patterns and analysis results obtained from the database, and the output is the fraud risk score.

[0998] Step 8:

[0999] The server generates a warning message based on the fraud risk. For example, it creates a warning message that reads, "This person is likely to be a fraudster. Avoid lending money to them and check for more information." The input is the fraud risk score, and the output is the generated warning message.

[1000] Step 9:

[1001] The server sends the generated warning message to the user terminal. The sent message is displayed on the user terminal. Specifically, the server sends a message to the user terminal using a message transmission protocol, and the user terminal displays the message on its screen. The input is the generated warning message, and the output is the warning message displayed on the user terminal.

[1002] Step 10:

[1003] The user provides feedback on the analysis results. This feedback is sent from the user terminal to the server. For example, the user may input "This warning was appropriate." The input is the feedback on the analysis results, and the output is the feedback data sent to the server.

[1004] Step 11:

[1005] The server receives user feedback and updates the machine learning model of the AI ​​analysis system. Specifically, the server passes the feedback data to the AI ​​analysis system and retrains the machine learning algorithm. The input is the feedback data, and the output is the updated machine learning model.

[1006] (Application example 1)

[1007] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1008] Users of communication platforms such as social networking sites and dating apps are at increased risk of becoming victims of fraud. Under these circumstances, there is a need for a method that automatically and accurately analyzes users' risk of fraud based on their past interactions and prevents fraud from occurring. However, with conventional methods, it is difficult for users to determine the risk themselves, which often results in delayed response. The present invention aims to solve this problem.

[1009] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1010] In this invention, the server includes: means for a user to input past interactions on the communication platform; means for transmitting the input data; means for the information processing device to input the data received from the user to a knowledge base analysis system; means for the knowledge base analysis system to analyze the data and extract elements based on fraud patterns; means for the information processing device to extract fraud patterns from the information accumulation system based on the analysis results; means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; means for notifying the user of the generated warning message; a mobile terminal having a user interface for copying and pasting interactions into an application; and a data transmission module having a protocol for encrypting and transmitting the input data. This enables a user to easily input past interactions and have their risks quickly and accurately evaluated.

[1011] "User" refers to any individual or organization using the communication platform.

[1012] A "communication platform" refers to an online service that allows users to exchange messages with each other, such as social networking sites or matching apps.

[1013] "Interaction" refers to the sending and receiving of messages between users on the communication platform.

[1014] "Input means" refers to a method or device that allows a user to provide the content of an interaction to the system in text form.

[1015] "Information processing device" refers to a computer device that receives, processes, and transmits data.

[1016] "Knowledge-based analysis system" refers to an artificial intelligence system for analyzing data based on patterns of fraudulent activity.

[1017] "Means for extracting elements" refers to the methods and devices that the knowledge-based analysis system uses to find features and patterns related to fraudulent activity in the data.

[1018] An "information collection system" refers to a database that accumulates and manages fraud patterns and past cases.

[1019] "Warning message" refers to a message intended to inform users of the risk of fraud.

[1020] "User interface" refers to the screens and input devices through which a user interacts with a system and inputs data.

[1021] "Mobile terminal" refers to an electronic device that a user can carry with them, such as a smartphone or tablet.

[1022] An "encryption and transmission protocol" refers to a method or communication protocol for encrypting and transmitting data to ensure secure data communication.

[1023] The system of the present invention allows users to input their past interactions on a communication platform, analyzes their hidden intentions using a knowledge-based analysis system, and assesses the risk of fraud. Specific embodiments are described below.

[1024] System Overview

[1025] The system mainly consists of the following components:

[1026] 1. User Device

[1027] A mobile device for inputting past interactions on a communication platform via a user interface. Users use a smartphone or tablet to input the relevant text into the system.

[1028] 2. Information Processing Device

[1029] It is a central computer system that receives data and executes the knowledge-based analysis system. The information processing device also manages the information collection system and notifies the results.

[1030] 3. Knowledge-based analysis system

[1031] The system analyzes data based on fraud patterns and extracts relevant elements. The knowledge-based analysis system uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[1032] 4. Information Collection System

[1033] This is a database for storing fraud patterns and past cases. The information processing device refers to this information accumulation system to assess fraud risk.

[1034] Processing Step Description

[1035] First, the user inputs a past conversation on the communication platform in text format, for example, "Hello! Would you like to meet up sometime? Can you lend me a little money first?" and pastes it into the application.

[1036] Next, the user terminal transmits the input data to the information processing device. This data transmission uses an encryption protocol such as SSL to ensure the security of the communication.

[1037] The information processing device passes the received data to a knowledge-based analysis system, which extracts elements that indicate fraud. The knowledge-based analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[1038] Based on the analysis results, the information processing device extracts fraud patterns from the information collection system and evaluates the fraud risk. For example, it calculates that the probability of fraud is 80%.

[1039] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and is sent to the user terminal.

[1040] Additionally, users can provide feedback on the analysis results, which is sent to the information processing device and used to improve the knowledge-based analysis system.

[1041] Specific examples

[1042] Example 1: Analysis of social media interactions

[1043] Type: "Hi! Want to meet up sometime? Can I borrow a little money first?"

[1044] Analysis result: 80% fraud risk

[1045] Notification message: "This person is likely a scammer. Avoid lending money and check for more information."

[1046] Example 2: Analysis of interactions on a dating app

[1047] Type: "I'd love to get married soon! I have an investment opportunity, would you be interested?"

[1048] Analysis result: 90% fraud risk

[1049] Notification message: "This person is very likely a scam. Be especially careful with investment proposals and check the full details."

[1050] Prompt Sentence Examples

[1051] "Analyze the following exchange and assess the risk of fraud: 'Hi! Would you like to meet up sometime? Can I borrow a little money first?'"

[1052] Analyze the following exchange and assess its risk of fraud: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[1053] In this way, users can analyze the true meaning of interactions on the communication platform and prevent themselves from falling victim to fraud.

[1054] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1055] Step 1:

[1056] A user inputs past interactions on a communication platform in text format. The user inputs (copy and paste) the text via the user interface of a smartphone or tablet, and prepares this text for subsequent analysis. The input includes content such as "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[1057] Step 2:

[1058] The user terminal transmits the input data to the information processing device. An encryption protocol such as SSL is used for data transmission to ensure communication security. The input in this step is the text entered by the user, and the output is the transmission of encrypted data.

[1059] Step 3:

[1060] The information processing device passes the received data to a knowledge-based analysis system. This data is decrypted in its original text form and provided to the knowledge-based analysis system for analysis. The input is the decrypted content of the encrypted data, and the output is analyzable text data.

[1061] Step 4:

[1062] A knowledge-based analysis system analyzes the data and extracts indicators of fraud. It uses a generative AI model trained on past fraud cases to identify phrases such as "lend me some money." The input is analyzable text data, and the output is a fraud risk score and indicators of fraud.

[1063] Step 5:

[1064] The data processor extracts fraud patterns from the data collection system based on the analysis results and assesses the fraud risk. The data collection system stores past fraud patterns and cases, and compares them with the current input data. The input is a fraud risk score and fraud indicator elements, and the output is a final fraud risk assessment.

[1065] Step 6:

[1066] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. The generated warning message includes content such as "This person is likely to be a fraudster. Avoid lending money and check the detailed information." The input is the fraud risk assessment, and the output is a warning message to the user.

[1067] Step 7:

[1068] The user provides feedback on the analysis results. The feedback is sent back to the information processing system, and the knowledge-based analysis system updates the model based on this feedback. The input in this step is the user's feedback, and the output is the updated analytical model.

[1069] This process allows users to quickly and automatically analyze whether their interactions on the communication platform pose a risk of fraud and receive appropriate warnings.

[1070] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1071] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, evaluates the possibility of fraud, and provides more accurate warnings by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments are described below.

[1072] System Overview

[1073] The system mainly consists of the following components:

[1074] 1. User Device

[1075] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[1076] 2. Server

[1077] It is the central computer system that receives data and runs the AI ​​analysis system and emotion engine. The server also manages the database and notifies the results.

[1078] 3. AI Analysis System

[1079] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[1080] 4. Emotion Engine

[1081] Analyzing emotions from user input data and classifying them into positive, negative, and neutral categories, the emotion engine provides emotional information to the AI ​​analysis system, which then adds it to the fraud risk assessment.

[1082] 5. Database

[1083] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[1084] Program processing explanation

[1085] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1086] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[1087] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[1088] At the same time, the server also passes the received data to the emotion engine, which analyzes the user's emotions. The emotion engine analyzes the emotions in the text and classifies them as positive, negative, or neutral.

[1089] Based on the analysis results, the server extracts fraudster patterns from the database and evaluates the fraud risk. Emotional information from the emotion engine is also added to the fraud risk assessment. For example, the probability of fraud is 80% and the user's emotion is determined to be negative.

[1090] The server generates a warning message based on the fraud risk assessment results and emotional information and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and sent to the user's terminal.

[1091] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model and emotion engine.

[1092] Specific examples

[1093] Example 1: Social media interactions

[1094] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1095] 2. The user terminal sends this text to the server.

[1096] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[1097] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as negative.

[1098] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[1099] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[1100] Example 2: Interactions on a dating app

[1101] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[1102] 2. The user terminal sends this text to the server.

[1103] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[1104] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as neutral.

[1105] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[1106] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[1107] In this way, users can analyze their interactions on social media and dating apps to understand their emotional state as well as the risk of fraud, enabling more accurate risk assessment and appropriate responses.

[1108] The processing flow will be explained below.

[1109] Step 1:

[1110] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1111] Step 2:

[1112] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[1113] Step 3:

[1114] The server receives the data sent from the user terminal and confirms that the data has been received.

[1115] Step 4:

[1116] The server cleanses the incoming data and formats it in a way that makes it easier to parse, for example by removing extra whitespace and special characters.

[1117] Step 5:

[1118] The server passes the cleansed data to the AI ​​analysis system, which then analyzes the data and extracts elements that indicate fraudulent activity, such as requests to lend money.

[1119] Step 6:

[1120] The server then passes the same cleansed data to a sentiment engine, which analyzes the data and identifies sentiments such as positive, negative, or neutral.

[1121] Step 7:

[1122] The emotion engine returns the analysis results to the server and determines whether the user has a negative emotion. For example, the emotion engine classifies the emotion as "negative."

[1123] Step 8:

[1124] The server extracts fraudster patterns from the database based on the results of the AI ​​analysis system and the emotion engine, and matches patterns that contain elements that indicate fraudulent activity.

[1125] Step 9:

[1126] The server comprehensively evaluates the fraud risk based on the analysis results and emotional information. For example, if the fraud rate is 80% and the user's emotional state is negative,

[1127] Step 10:

[1128] The server generates a warning message based on the fraud risk assessment results and emotional information. For example, it creates a message saying, "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[1129] Step 11:

[1130] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[1131] Step 12:

[1132] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[1133] Step 13:

[1134] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[1135] Step 14:

[1136] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model and emotion engine in the future.

[1137] This is the specific processing flow of this system. As a result, users can detect fraudulent behavior early through their interactions on social media or dating apps, as well as recognize their own emotional state and take appropriate action.

[1138] Example 2

[1139] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1140] In conventional social networking services and dating applications, there was no system that effectively assessed and warned users about their risk of being scammed. As a result, users were more likely to fall prey to scammers' sophisticated tactics, increasing the risk of financial loss. In addition, the accuracy of warnings was low because the system did not take into account the user's emotional state.

[1141] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to input past interactions on a social networking service or a matching application; a means for transmitting the input data; a means for the processing device to input data received from the user to an AI analysis system; a means for the AI ​​analysis system to analyze the data and extract elements based on fraud patterns; a means for the processing device to extract fraud patterns from the data management system based on the analysis results; a means for the processing device to evaluate fraud risk based on the analysis results and generate a warning message; a means for notifying the user of the generated warning message; and a means for analyzing emotions in the input data and adding emotion-based information to the fraud risk evaluation. This allows users to quickly and accurately evaluate the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[1142] "User" refers to an individual who uses a social networking service or dating application.

[1143] "Social networking service" refers to an online platform that enables users to communicate with other individuals.

[1144] "Matching Application" means a mobile or web application that enables users to meet and connect with other individuals.

[1145] "Past interactions" refers to the history of messages exchanged by a user on social networking services or matching applications.

[1146] "Means for input" refers to an interface or device that allows a user to send past interactions to the system in text form.

[1147] "Processing device" refers to a central computer system for receiving, analyzing, and evaluating input data.

[1148] "Artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze data and detect patterns of fraudulent activity.

[1149] A "data management system" refers to an information storage system that stores past fraud cases and patterns and references them as needed.

[1150] "Fraud" refers to any dishonest activity that attempts to deceive users into giving them money or personal information.

[1151] "Elements" refer to the characteristics extracted by artificial intelligence analysis systems to identify patterns of fraudulent activity.

[1152] "Analysis results" refers to the conclusions and data derived by the AI ​​analysis system through data analysis.

[1153] "Fraud risk" refers to the degree of likelihood that the input data represents fraudulent activity.

[1154] "Warning Message" refers to a message containing a warning provided to a user based on a fraud risk assessment.

[1155] "Means of notification" refers to the communication means or interface used to convey a warning message to the user.

[1156] "Means for analyzing emotions" refers to systems or algorithms for extracting emotional information from text data and analyzing that information.

[1157] "Emotion-based information" refers to data that represents a user's emotional state and is used in fraud risk assessment.

[1158] This invention provides a system that inputs past interactions a user has had on social networking services or matching applications, analyzes the data using an AI analysis system and a sentiment analysis engine, and assesses the risk of fraud. This system is comprised of a user terminal, a server, an AI analysis system, a sentiment analysis engine, and a database.

[1159] Hardware and software used

[1160] User device: The smartphone or computer that the user uses to enter data and receive results. Examples: iPhone, Android device, Windows PC, Mac.

[1161] Server: A high-performance computer system to receive data and perform analytical processing. Examples: AWS EC2, Google Cloud Compute Engine.

[1162] Communication Protocol: The secure communication method used to send and receive data. For example: SSL / TLS.

[1163] Artificial intelligence analysis systems: Use machine learning algorithms that learn fraud patterns. Examples: Scikit-learn, TensorFlow.

[1164] Sentiment analysis engine: uses algorithms that analyze user sentiment. Example: IBM Watson Natural Language Understanding.

[1165] Database: A system for storing information on past fraud cases and patterns. Examples: MySQL, PostgreSQL.

[1166] System Overview

[1167] 1. User enters data:

[1168] The user inputs past interactions in text format, for example, "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[1169] 2. Data transmission:

[1170] The user terminal sends the entered data to the server using an encryption protocol (such as SSL / TLS).

[1171] 3. Data Receipt and Analysis:

[1172] The server receives the data and first analyzes it with an artificial intelligence analysis system, which uses machine learning algorithms that have learned fraudulent patterns to extract elements, such as phrases like "lend me some money."

[1173] 4. Emotion analysis:

[1174] The data is passed to a sentiment analysis engine to analyze the user's emotional state. The text is classified as positive, negative, or neutral. For example, if the user's message is offensive, it will be classified as negative.

[1175] 5. Fraud risk assessment and warning message generation:

[1176] The server integrates the results of the AI ​​analysis system and the sentiment analysis engine, and compares them with past fraud patterns in the database to assess the fraud risk. For example, if the fraud risk is determined to be high (80%), a warning message is generated based on that risk.

[1177] 6. Warning message notification:

[1178] The server generates a warning message and sends it to the user's device, which displays the message "This person is likely a fraud. Avoid lending money to them and check for more information."

[1179] 7. Gathering Feedback:

[1180] Users provide feedback on the analysis results, which is sent to the server and used to improve the artificial intelligence model and sentiment analysis engine.

[1181] Specific examples

[1182] Example 1: Social media interactions

[1183] 1. Users copy social media conversations and enter them into the system.

[1184] For example: "Hi! Would you like to meet up sometime? Can I borrow a little money first?"

[1185] 2. The user terminal sends this text to the server.

[1186] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[1187] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as negative.

[1188] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[1189] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[1190] Example 2: Interactions on a dating app

[1191] 1. The user copies the conversations they have had on the dating app and enters them into the system.

[1192] Example: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[1193] 2. The user terminal sends this text to the server.

[1194] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[1195] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as neutral.

[1196] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[1197] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[1198] This allows users to quickly and accurately assess the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[1199] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1200] Step 1:

[1201] The user inputs past interactions in text format on social networking services or dating applications. Specifically, the user inputs past messages by copying and pasting them into the application screen on their smartphone or computer. An example of input is the message "Hello! Would you like to meet sometime? Can you lend me a little money first?" The input data is saved in text format.

[1202] Step 2:

[1203] The user terminal sends the entered text data to the server. At this time, an encryption protocol such as SSL / TLS is used to ensure the security of the communication. The input is text data, and the output is encrypted data. In concrete terms, the application on the user terminal issues a send request to the server, and a progress bar is displayed.

[1204] Step 3:

[1205] The server passes the received data to an artificial intelligence analysis system, which extracts elements that indicate fraud. The data input is encrypted text data, and the output contains elements related to fraud patterns (e.g., keywords such as "money" and "lend me"). This processing uses machine learning algorithms (e.g., Scikit-learn, TensorFlow). Specifically, the server records a "receiving data" log and displays an analysis progress bar in percentage.

[1206] Step 4:

[1207] At the same time, the server passes the received data to a sentiment analysis engine to analyze the user's sentiment. The data input is the original text data, and the output is the sentiment classification result (positive, negative, neutral). A sentiment analysis algorithm (e.g., IBM Watson Natural Language Understanding) is used for sentiment analysis. Specifically, the server records a log of "sentiment analysis in progress" and displays the progress of sentiment analysis in percentage.

[1208] Step 5:

[1209] The server integrates the results of the AI ​​analysis system and sentiment analysis, and compares them with past fraud patterns in the database to assess the fraud risk. The input is the elements of the fraud pattern and the sentiment classification results, and the output is the fraud risk assessment result (e.g., fraud risk 80%). The server generates a warning message based on the analysis results. Specifically, the server records a log of "fraud risk assessment in progress" and displays the fraud risk value.

[1210] Step 6:

[1211] The server generates a warning message and sends it to the user's device. The input is the fraud risk assessment result, and the output is a warning message (e.g., "This person is likely to be a fraudster. Avoid lending money and check for more information."). Specifically, the server issues a message sending request to the user's device, and a notification is displayed on the user's device screen.

[1212] Step 7:

[1213] The user provides feedback on the system's analysis results. The input is the user's feedback (e.g., "This analysis result is accurate"), and the output is the feedback data recorded on the server. Specifically, the user enters a comment using the feedback form and clicks the submit button. The server receives the feedback and records it in a log.

[1214] The above steps enable highly accurate fraud risk assessment and the generation of warning messages, allowing users to safely use social networking sites and dating applications.

[1215] (Application example 2)

[1216] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1217] In modern society, fraudulent activities via social networking sites and other information exchange platforms are on the rise, potentially causing financial losses to many people. Furthermore, fraudulent methods are becoming increasingly sophisticated, making it difficult for users to assess the risks involved in their own communications. Therefore, there is a need for a system that can detect these fraudulent activities early and warn users. The purpose of this invention is to solve these problems and ensure user safety.

[1218] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the user's emotions and adding an emotional evaluation to the warning message, means for notifying the user of the generated warning message, and means for generating a prompt sentence using a framework to evaluate the probability of fraud and the user's emotional state. This makes it possible to accurately evaluate the fraud risk and provide an appropriate warning that takes the user's emotional state into account in real time.

[1219] "User" refers to an entity (generally a human being) that uses a means of information exchange.

[1220] "Information exchange means" refers to systems that include software and hardware that allow users to communicate, such as social networking sites and communication apps.

[1221] "Central Computing Unit" means the central computer system that receives, analyzes, and stores data.

[1222] An "artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze input data and extract specific patterns or elements.

[1223] A "database" refers to an information management system that references and stores past data and specific patterns based on analysis results.

[1224] "Fraud" refers to any dishonest act that involves deceiving others to obtain money or information.

[1225] "Emotion analysis" refers to the technology of determining a user's emotional state from data such as text and voice.

[1226] A "prompt" is an instruction given to a generative AI model, a piece of text that provides context for a specific analysis or generation task.

[1227] An "encryption protocol" is a technology for ensuring the security of communications when transmitting data, encrypting the data to protect it from unauthorized access by third parties.

[1228] "Warning Message" means a message intended to inform users of fraud risks or other important information.

[1229] The system of the present invention analyzes past interactions of users of social networking sites and information exchange means, assesses fraud risk and emotional state, and issues warnings. This system is composed of a user terminal, a central computing device, an AI analysis system, a database, an emotion analysis engine, and a prompt sentence generation means.

[1230] System Overview

[1231] 1. User Device

[1232] Users can use their smartphones or computers to input past interactions on social media or communication apps. For example, they can copy conversations on a chat app and input them into the system.

[1233] 2. Central computing device

[1234] This is the central computer system that receives data and runs the AI ​​analysis system and sentiment analysis engine. It also manages the database and notifies results.

[1235] 3. Artificial Intelligence Analysis System

[1236] It uses machine learning algorithms to study past fraud cases and extract elements based on fraud patterns with high accuracy, identifying phrases and patterns that indicate fraudulent activity and assessing fraud risk.

[1237] 4. Sentiment Analysis Engine

[1238] Analyzes sentiment from user input data and classifies it into positive, negative, and neutral categories, providing additional information for fraud risk assessment.

[1239] 5. Database

[1240] It is an information collection system that stores patterns and past cases of fraudulent activity. The central computing device refers to this database to assess fraud risk.

[1241] 6. Prompt generation method

[1242] The framework generates prompts to assess the probability of fraud and the emotional state of the user. The generated prompts are then fed into an AI model for specific analysis and evaluation.

[1243] Program processing explanation

[1244] The central computing unit passes the data entered by the user to the AI ​​analysis system, which uses machine learning algorithms to analyze the data and assess the risk of fraud. At the same time, the emotion analysis engine analyzes the user's emotions and classifies them as positive, negative, or neutral. Based on the analysis results and the emotion assessment, the central computing unit generates a warning message and notifies the user's device. It also generates a prompt sentence including the emotion assessment and inputs it into the AI ​​model to improve the accuracy of the analysis.

[1245] Specific examples

[1246] Example 1: Social media interactions

[1247] A user copies a conversation on social media and inputs it into the system. The content is, "Hello! Would you like to meet sometime? Can you lend me a little money first?" The central computing device passes this text to the artificial intelligence analysis system, which extracts elements that indicate fraud. At the same time, the emotion analysis engine analyzes the user's emotional state and classifies it as negative. Finally, the central computing device sends a warning message to the user's device saying, "This person is likely to be a fraudster. Be careful not to lend them money."

[1248] Prompt Sentence Examples

[1249] "What is the fraud risk of this message? We use a fraud risk assessment model."

[1250] "What is the emotional state of this message? We use sentiment analysis models."

[1251] In this way, users can analyze past interactions, understand fraud risk and emotional state, and make risk assessments and take appropriate measures.

[1252] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1253] Step 1:

[1254] A user copies past interactions on social networking sites or other information exchange platforms and inputs them into their device. The input data is in text format. This text becomes input data that is later sent from the device to the central computing device for analysis.

[1255] Step 2:

[1256] The terminal transmits the input data to the central computing device, where it is encrypted using a secure communication protocol (e.g., SSL) to protect it from unauthorized access by third parties. The output is the encrypted text data received at the central computing device.

[1257] Step 3:

[1258] The central computing unit decodes the received data and passes it to the AI ​​analysis system. The input is the decoded text data, which is preprocessed for analysis. Preprocessing includes normalizing the data and removing unnecessary characters. The output is data converted into a form suitable for analysis.

[1259] Step 4:

[1260] An AI analysis system analyzes text data and extracts elements that indicate fraud. It uses a model trained on past fraud data to assess the fraud risk from the input data. The output is a fraud risk assessment result, specifically a fraud likelihood score and a list of related elements.

[1261] Step 5:

[1262] The central computing device passes the fraud risk assessment result to the emotion analysis engine, which analyzes the user's emotions. The input is the fraud risk assessment result and text data, which the emotion analysis engine analyzes to determine the emotional state (positive, negative, neutral). The output is the emotion assessment result.

[1263] Step 6:

[1264] The central computing device integrates the fraud risk assessment results and the emotion assessment results to generate a warning message. The input is the fraud risk score and the emotion state information, and based on these, it generates an appropriate warning message for the user. The output is the warning message.

[1265] Step 7:

[1266] The generated warning message is sent to the user terminal. The central computing device sends the warning message to the user terminal, and the user receives the warning message so that the user can take appropriate action regarding the fraud risk. The output is the warning message received by the user.

[1267] Step 8:

[1268] The central computing device uses the framework to generate prompt sentences for assessing the probability of fraud and the emotional state. The inputs are the fraud risk assessment results and the emotional assessment results, and specific prompt sentences are generated based on these. The output is the prompt sentence.

[1269] Step 9:

[1270] The generated prompt sentences are input into an artificial intelligence model to improve analysis accuracy. Based on the prompt sentences, additional analysis or model retraining is performed as needed. The output is an improved analysis model or new analysis results.

[1271] In this way, at each step, the terminal, central computing device, artificial intelligence analysis system, and emotion analysis engine work together to assess fraud risk and emotional state and provide appropriate warnings.

[1272] 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.

[1273] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1274] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1275] [Fourth embodiment]

[1276] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1277] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1278] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1279] 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.

[1280] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1281] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1282] 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.

[1283] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1284] 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.

[1285] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1286] 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.

[1287] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1288] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1289] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, and evaluates the possibility of fraud. Specific embodiments are described below.

[1290] System Overview

[1291] The system mainly consists of the following components:

[1292] 1. User Device

[1293] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[1294] 2. Server

[1295] The server is the central computer system that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[1296] 3. AI Analysis System

[1297] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[1298] 4. Database

[1299] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[1300] Program processing explanation

[1301] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1302] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[1303] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[1304] Based on the analysis results, the server extracts fraudster patterns from the database and assesses the fraud risk, for example, calculating that the probability of fraud is 80%.

[1305] The server generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check for more information." and is sent to the user's device.

[1306] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model.

[1307] Specific examples

[1308] Example 1: Social media interactions

[1309] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1310] 2. The user terminal sends this text to the server.

[1311] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[1312] 4. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[1313] 5. The server creates a warning message saying, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[1314] Example 2: Interactions on a dating app

[1315] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[1316] 2. The user terminal sends this text to the server.

[1317] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[1318] 4. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[1319] 5. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[1320] In this way, users can analyze the true meaning of interactions on social media and dating apps and prevent themselves from falling victim to fraud.

[1321] The processing flow will be explained below.

[1322] Step 1:

[1323] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1324] Step 2:

[1325] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[1326] Step 3:

[1327] The server receives the data sent from the user terminal and confirms that the data has been received.

[1328] Step 4:

[1329] Before the server can pass the received data to the AI ​​analysis system, it cleanses the text data and formats it in a way that makes it easier to analyze, for example by removing unnecessary spaces and special characters.

[1330] Step 5:

[1331] An AI analysis system analyzes incoming data and extracts elements based on scammer patterns, specifically identifying phrases that indicate fraudulent activity, such as "please lend me some money."

[1332] Step 6:

[1333] Based on the results of the AI ​​analysis system, the server extracts fraudster patterns from a database that records past fraud cases and typical patterns of fraudulent behavior.

[1334] Step 7:

[1335] The server will evaluate multiple factors (e.g., phrases used, frequency of interactions, inconsistencies, etc.) and calculate a fraud risk as a percentage. For example, it may determine that the fraud risk is 80%.

[1336] Step 8:

[1337] The server generates a warning message based on the fraud risk assessment result, such as "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[1338] Step 9:

[1339] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[1340] Step 10:

[1341] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[1342] Step 11:

[1343] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[1344] Step 12:

[1345] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model in the future.

[1346] This is the specific processing flow of this system, which allows users to detect in advance the risk of becoming a victim of fraud through interactions on social media or dating apps, and respond appropriately.

[1347] Example 1

[1348] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1349] In modern society, fraudulent activities involving the misuse of communication applications are becoming more common. Users of social networking sites and dating apps are particularly susceptible to the sophisticated tactics of fraudsters. In addition to standard security measures, there is a demand for systems that can accurately detect potential fraudulent activity and quickly warn users. However, conventional systems have difficulty efficiently detecting signs of fraud and incorporating user feedback to improve accuracy. The purpose of this invention is to solve these problems.

[1350] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1351] In this invention, the server includes: a means for a user to input past interactions in a communication application; a means for transmitting the input data to an information processing device; a means for the information processing device to input the data received from the user to a machine learning model; a means for the machine learning model to analyze the data and extract elements based on fraudulent activity patterns; a means for the information processing device to extract fraudulent activity patterns from a database based on the analysis results; a means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; and a means for notifying the user of the generated warning message. This enables the system to detect possible fraudulent activity with high accuracy and quickly issue a warning to the user. Furthermore, by adding a means for updating the machine learning model based on user feedback, the system can continuously improve its accuracy. As a result, users can use communication applications with peace of mind.

[1352] "User" refers to an individual or corporation that uses this system.

[1353] "Communication applications" refers to all software that allows messages and information to be exchanged over the Internet, such as social networking sites and dating apps.

[1354] "Past interactions" refers to the history of messages and chats that a user has had on a communication application.

[1355] "Information processing device" refers to a general term for a computer system that receives, analyzes, stores, and notifies data.

[1356] A "machine learning model" refers to an algorithm that uses artificial intelligence to learn patterns from past data and analyze and predict new data.

[1357] "Fraud patterns" refer to common behavioral characteristics or phrases extracted from past instances of fraud or malicious behavior.

[1358] "Analysis results" refer to the indicators and elements of fraudulent activity extracted by the machine learning model from the input data.

[1359] A "database" refers to a system that stores information necessary for analysis, such as patterns of fraudulent activity and past cases.

[1360] "Fraud risk" refers to the possibility of fraud or malicious intent calculated based on the analysis results.

[1361] The "warning message" refers to information sent to the user to call their attention based on the analysis results.

[1362] "Feedback" refers to the opinions and evaluations provided by users regarding the system's analysis results.

[1363] MODE FOR CARRYING OUT THE INVENTION

[1364] The present invention is a system that allows users to input their past interactions on social media or dating apps, analyzes that data with an AI analysis system, evaluates the possibility of fraud, and issues a warning to the user. The program processing of this system is explained in detail below.

[1365] System configuration

[1366] This invention mainly consists of the following hardware and software:

[1367] 1. User Device

[1368] A device that allows a user to input past interactions with a communication application, including a smartphone or computer.

[1369] 2. Server

[1370] The server is the central information processing device that receives data and runs the AI ​​analysis system. It also manages the database and notifies users of the results.

[1371] 3. AI Analysis System

[1372] The AI ​​analysis system analyzes data based on fraud patterns and extracts elements of fraudulent activity. It uses machine learning algorithms to enable highly accurate data analysis. For example, it uses machine learning frameworks such as TensorFlow and PyTorch.

[1373] 4. Database

[1374] It is an information collection system that stores fraud patterns and past cases. The server refers to this database to assess fraud risk.

[1375] Program processing overview

[1376] First, the user inputs a past conversation in a communication application in text format. For example, the user might input something like, "Hello! Would you like to meet sometime? Can you lend me a little money first?" into the system.

[1377] The user terminal then sends the entered data to the server, using an encryption protocol such as SSL to ensure the security of the communication.

[1378] The server passes the received data to an AI analysis system, which extracts elements that indicate fraudulent activity. The AI ​​analysis system learns from past cases of fraud and identifies and analyzes phrases such as "lend me some money."

[1379] Based on the analysis results, the server extracts fraudulent patterns from the database and assesses the fraud risk. For example, if the probability of fraud is calculated to be 80%, it generates a specific warning message to inform the user of the risk.

[1380] The generated warning message may say something like, "This person is likely to be a fraud. Avoid lending money to them and check for more information," and is sent from the server to the user's terminal.

[1381] Users can provide feedback on the analysis results. This feedback is sent to the server and used to improve the AI ​​model. For example, the feedback could be, "This warning was appropriate."

[1382] Specific examples

[1383] Example 1: Social media interactions

[1384] 1. The user enters the chat history from a social networking site into the system: "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[1385] 2. The user terminal sends this text to the server.

[1386] 3. The server passes the received data to the AI ​​analysis system.

[1387] 4. The server's AI analysis system determines that the phrase "lend me money" is 80% likely to be a scam.

[1388] 5. Based on the results, the server generates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and notifies the user.

[1389] 6. The user sends feedback saying, "This warning was appropriate."

[1390] 7. The server stores the feedback in a database and uses it for future analysis.

[1391] Example 2: Interactions on a dating app

[1392] 1. The user enters the chat history from the dating app into the system: "I want to get married soon! There's an investment opportunity. Are you interested?"

[1393] 2. The user device sends this data to the server.

[1394] 3. The server passes the received data to the AI ​​analysis system.

[1395] 4. The server's AI analysis system detects phrases related to "investment opportunities" and rates the likelihood of fraud at 90%.

[1396] 5. Based on the high fraud risk, the server generates a warning message stating, "This person is very likely to be a fraud. Please be especially careful with investment proposals and check the detailed information." and notifies the user.

[1397] 6. The user sends feedback saying, "This warning was helpful."

[1398] 7. The server stores the feedback in a database and uses it to improve the AI ​​model.

[1399] In this way, the system of the present invention can analyze the true intent of users' interactions in communication applications and prevent fraud before it occurs. Furthermore, by incorporating user feedback, the system's analysis accuracy can be continuously improved.

[1400] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1401] Step 1:

[1402] The user inputs past conversations in a communication application in text format. For example, the user copies and pastes the chat content from an SNS such as "Hello! Let's meet sometime. Can you lend me a little money first?" into the system. The input text is temporarily stored on the user's device.

[1403] Step 2:

[1404] The user terminal sends the entered text data to the server. At this time, security is ensured when sending data using encryption protocols such as SSL. Specifically, when the user presses the "Send" button, the text data is encrypted and sent to the server via the Internet. The input is the text data entered by the user, and the output is the encrypted data received by the server.

[1405] Step 3:

[1406] The server receives the data sent from the user terminal. The received data is decrypted and restored to the original text data. Specifically, the decryption program on the server decrypts the encrypted data according to the SSL protocol to obtain plain text data. The input is encrypted text data, and the output is decrypted text data.

[1407] Step 4:

[1408] The server passes the decrypted text data to the AI ​​analysis system. The passed data is added to a processing queue within the AI ​​analysis system. Specifically, the server converts the text data into a format suitable for the machine learning model and sends it to the analysis system. The input is the decrypted text data, and the output is the data input to the machine learning model.

[1409] Step 5:

[1410] The AI ​​analysis system analyzes the data and extracts elements that indicate fraudulent activity. For example, it identifies the phrase "lend me some money" and extracts corresponding patterns. The input is the text data to be analyzed, and the output is the extracted elements of fraudulent activity.

[1411] Step 6:

[1412] The server then references the database based on the analysis results and compares them with past fraudulent activity patterns. Specifically, the server executes a database query to search for relevant past cases. The input is the extracted fraudulent activity elements, and the output is the corresponding fraudulent activity patterns retrieved from the database.

[1413] Step 7:

[1414] The server assesses the fraud risk based on the fraudulent activity patterns and analysis results. This assessment is converted into a numerical value according to an algorithm, for example, the probability of fraud is calculated as 80%. Specifically, the server executes the risk assessment algorithm and calculates the fraud risk score. The input is the fraudulent activity patterns and analysis results obtained from the database, and the output is the fraud risk score.

[1415] Step 8:

[1416] The server generates a warning message based on the fraud risk. For example, it creates a warning message that reads, "This person is likely to be a fraudster. Avoid lending money to them and check for more information." The input is the fraud risk score, and the output is the generated warning message.

[1417] Step 9:

[1418] The server sends the generated warning message to the user terminal. The sent message is displayed on the user terminal. Specifically, the server sends a message to the user terminal using a message transmission protocol, and the user terminal displays the message on its screen. The input is the generated warning message, and the output is the warning message displayed on the user terminal.

[1419] Step 10:

[1420] The user provides feedback on the analysis results. This feedback is sent from the user terminal to the server. For example, the user may input "This warning was appropriate." The input is the feedback on the analysis results, and the output is the feedback data sent to the server.

[1421] Step 11:

[1422] The server receives user feedback and updates the machine learning model of the AI ​​analysis system. Specifically, the server passes the feedback data to the AI ​​analysis system and retrains the machine learning algorithm. The input is the feedback data, and the output is the updated machine learning model.

[1423] (Application example 1)

[1424] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1425] Users of communication platforms such as social networking sites and dating apps are at increased risk of becoming victims of fraud. Under these circumstances, there is a need for a method that automatically and accurately analyzes users' risk of fraud based on their past interactions and prevents fraud from occurring. However, with conventional methods, it is difficult for users to determine the risk themselves, which often results in delayed response. The present invention aims to solve this problem.

[1426] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1427] In this invention, the server includes: means for a user to input past interactions on the communication platform; means for transmitting the input data; means for the information processing device to input the data received from the user to a knowledge base analysis system; means for the knowledge base analysis system to analyze the data and extract elements based on fraud patterns; means for the information processing device to extract fraud patterns from the information accumulation system based on the analysis results; means for the information processing device to evaluate fraud risks based on the analysis results and generate a warning message; means for notifying the user of the generated warning message; a mobile terminal having a user interface for copying and pasting interactions into an application; and a data transmission module having a protocol for encrypting and transmitting the input data. This enables a user to easily input past interactions and have their risks quickly and accurately evaluated.

[1428] "User" refers to any individual or organization using the communication platform.

[1429] A "communication platform" refers to an online service that allows users to exchange messages with each other, such as social networking sites or matching apps.

[1430] "Interaction" refers to the sending and receiving of messages between users on the communication platform.

[1431] "Input means" refers to a method or device that allows a user to provide the content of an interaction to the system in text form.

[1432] "Information processing device" refers to a computer device that receives, processes, and transmits data.

[1433] "Knowledge-based analysis system" refers to an artificial intelligence system for analyzing data based on patterns of fraudulent activity.

[1434] "Means for extracting elements" refers to the methods and devices that the knowledge-based analysis system uses to find features and patterns related to fraudulent activity in the data.

[1435] An "information collection system" refers to a database that accumulates and manages fraud patterns and past cases.

[1436] "Warning message" refers to a message intended to inform users of the risk of fraud.

[1437] "User interface" refers to the screens and input devices through which a user interacts with a system and inputs data.

[1438] "Mobile terminal" refers to an electronic device that a user can carry with them, such as a smartphone or tablet.

[1439] An "encryption and transmission protocol" refers to a method or communication protocol for encrypting and transmitting data to ensure secure data communication.

[1440] The system of the present invention allows users to input their past interactions on a communication platform, analyzes their hidden intentions using a knowledge-based analysis system, and assesses the risk of fraud. Specific embodiments are described below.

[1441] System Overview

[1442] The system mainly consists of the following components:

[1443] 1. User Device

[1444] A mobile device for inputting past interactions on a communication platform via a user interface. Users use a smartphone or tablet to input the relevant text into the system.

[1445] 2. Information Processing Device

[1446] It is a central computer system that receives data and executes the knowledge-based analysis system. The information processing device also manages the information collection system and notifies the results.

[1447] 3. Knowledge-based analysis system

[1448] The system analyzes data based on fraud patterns and extracts relevant elements. The knowledge-based analysis system uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[1449] 4. Information Collection System

[1450] This is a database for storing fraud patterns and past cases. The information processing device refers to this information accumulation system to assess fraud risk.

[1451] Processing Step Description

[1452] First, the user inputs a past conversation on the communication platform in text format, for example, "Hello! Would you like to meet up sometime? Can you lend me a little money first?" and pastes it into the application.

[1453] Next, the user terminal transmits the input data to the information processing device. This data transmission uses an encryption protocol such as SSL to ensure the security of the communication.

[1454] The information processing device passes the received data to a knowledge-based analysis system, which extracts elements that indicate fraud. The knowledge-based analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[1455] Based on the analysis results, the information processing device extracts fraud patterns from the information collection system and evaluates the fraud risk. For example, it calculates that the probability of fraud is 80%.

[1456] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and is sent to the user terminal.

[1457] Additionally, users can provide feedback on the analysis results, which is sent to the information processing device and used to improve the knowledge-based analysis system.

[1458] Specific examples

[1459] Example 1: Analysis of social media interactions

[1460] Type: "Hi! Want to meet up sometime? Can I borrow a little money first?"

[1461] Analysis result: 80% fraud risk

[1462] Notification message: "This person is likely a scammer. Avoid lending money and check for more information."

[1463] Example 2: Analysis of interactions on a dating app

[1464] Type: "I'd love to get married soon! I have an investment opportunity, would you be interested?"

[1465] Analysis result: 90% fraud risk

[1466] Notification message: "This person is very likely a scam. Be especially careful with investment proposals and check the full details."

[1467] Prompt Sentence Examples

[1468] "Analyze the following exchange and assess the risk of fraud: 'Hi! Would you like to meet up sometime? Can I borrow a little money first?'"

[1469] Analyze the following exchange and assess its risk of fraud: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[1470] In this way, users can analyze the true meaning of interactions on the communication platform and prevent themselves from falling victim to fraud.

[1471] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1472] Step 1:

[1473] A user inputs past interactions on a communication platform in text format. The user inputs (copy and paste) the text via the user interface of a smartphone or tablet, and prepares this text for subsequent analysis. The input includes content such as "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[1474] Step 2:

[1475] The user terminal transmits the input data to the information processing device. An encryption protocol such as SSL is used for data transmission to ensure communication security. The input in this step is the text entered by the user, and the output is the transmission of encrypted data.

[1476] Step 3:

[1477] The information processing device passes the received data to a knowledge-based analysis system. This data is decrypted in its original text form and provided to the knowledge-based analysis system for analysis. The input is the decrypted content of the encrypted data, and the output is analyzable text data.

[1478] Step 4:

[1479] A knowledge-based analysis system analyzes the data and extracts indicators of fraud. It uses a generative AI model trained on past fraud cases to identify phrases such as "lend me some money." The input is analyzable text data, and the output is a fraud risk score and indicators of fraud.

[1480] Step 5:

[1481] The data processor extracts fraud patterns from the data collection system based on the analysis results and assesses the fraud risk. The data collection system stores past fraud patterns and cases, and compares them with the current input data. The input is a fraud risk score and fraud indicator elements, and the output is a final fraud risk assessment.

[1482] Step 6:

[1483] The information processing device generates a warning message based on the fraud risk assessment result and notifies the user. The generated warning message includes content such as "This person is likely to be a fraudster. Avoid lending money and check the detailed information." The input is the fraud risk assessment, and the output is a warning message to the user.

[1484] Step 7:

[1485] The user provides feedback on the analysis results. The feedback is sent back to the information processing system, and the knowledge-based analysis system updates the model based on this feedback. The input in this step is the user's feedback, and the output is the updated analytical model.

[1486] This process allows users to quickly and automatically analyze whether their interactions on the communication platform pose a risk of fraud and receive appropriate warnings.

[1487] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1488] The system of the present invention allows users to input their past interactions on social media or dating apps, analyzes their hidden intentions using an AI analysis system, evaluates the possibility of fraud, and provides more accurate warnings by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments are described below.

[1489] System Overview

[1490] The system mainly consists of the following components:

[1491] 1. User Device

[1492] It is a device for inputting messages sent on social media or dating apps. Users use their smartphones or computers to input the relevant text into the system.

[1493] 2. Server

[1494] It is the central computer system that receives data and runs the AI ​​analysis system and emotion engine. The server also manages the database and notifies the results.

[1495] 3. AI Analysis System

[1496] The AI ​​analysis system analyzes data based on fraudster patterns and extracts relevant elements. It uses machine learning algorithms to learn from past fraud cases, enabling highly accurate analysis.

[1497] 4. Emotion Engine

[1498] Analyzing emotions from user input data and classifying them into positive, negative, and neutral categories, the emotion engine provides emotional information to the AI ​​analysis system, which then adds it to the fraud risk assessment.

[1499] 5. Database

[1500] It is an information collection system that stores fraudster patterns and past cases. The server refers to this database to assess fraud risk.

[1501] Program processing explanation

[1502] First, the user enters past interactions on social media or a dating app in text format. For example, they might enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1503] Next, the user terminal sends the entered data to the server, using encryption protocols such as SSL to ensure communication security.

[1504] The server passes the received data to an AI analysis system that extracts elements that indicate fraud. The AI ​​analysis system has learned from past fraud cases and can identify and analyze phrases such as "lend me some money."

[1505] At the same time, the server also passes the received data to the emotion engine, which analyzes the user's emotions. The emotion engine analyzes the emotions in the text and classifies them as positive, negative, or neutral.

[1506] Based on the analysis results, the server extracts fraudster patterns from the database and evaluates the fraud risk. Emotional information from the emotion engine is also added to the fraud risk assessment. For example, the probability of fraud is 80% and the user's emotion is determined to be negative.

[1507] The server generates a warning message based on the fraud risk assessment results and emotional information and notifies the user. Specifically, a message is generated stating, "This person is likely to be a fraudster. Avoid lending money and check the detailed information." and sent to the user's terminal.

[1508] Additionally, users can provide feedback on the analysis results, which is sent to the server and used to improve the AI ​​model and emotion engine.

[1509] Specific examples

[1510] Example 1: Social media interactions

[1511] 1. A user copies a conversation on a social networking site and enters it into the system. The content is something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1512] 2. The user terminal sends this text to the server.

[1513] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[1514] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as negative.

[1515] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[1516] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[1517] Example 2: Interactions on a dating app

[1518] 1. A user copies the conversation they had on a dating app and enters it into the system. They enter something like, "I want to get married soon! There's an investment opportunity. Are you interested?"

[1519] 2. The user terminal sends this text to the server.

[1520] 3. The server passes the received data to an AI analysis system, which extracts elements that indicate fraud.

[1521] 4. At the same time, the server passes the received data to the emotion engine to analyze the user's emotion. The emotion engine classifies the user's emotion as neutral.

[1522] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[1523] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[1524] In this way, users can analyze their interactions on social media and dating apps to understand their emotional state as well as the risk of fraud, enabling more accurate risk assessment and appropriate responses.

[1525] The processing flow will be explained below.

[1526] Step 1:

[1527] The user enters past interactions on social media or dating apps in text format into the application's input field. For example, they enter something like, "Hello! Would you like to meet up sometime? Can you lend me a little money first?"

[1528] Step 2:

[1529] The user's device sends the entered text data to the server, and encryption protocols such as SSL are used to ensure the security of the data transmission.

[1530] Step 3:

[1531] The server receives the data sent from the user terminal and confirms that the data has been received.

[1532] Step 4:

[1533] The server cleanses the incoming data and formats it in a way that makes it easier to parse, for example by removing extra whitespace and special characters.

[1534] Step 5:

[1535] The server passes the cleansed data to the AI ​​analysis system, which then analyzes the data and extracts elements that indicate fraudulent activity, such as requests to lend money.

[1536] Step 6:

[1537] The server then passes the same cleansed data to a sentiment engine, which analyzes the data and identifies sentiments such as positive, negative, or neutral.

[1538] Step 7:

[1539] The emotion engine returns the analysis results to the server and determines whether the user has a negative emotion. For example, the emotion engine classifies the emotion as "negative."

[1540] Step 8:

[1541] The server extracts fraudster patterns from the database based on the results of the AI ​​analysis system and the emotion engine, and matches patterns that contain elements that indicate fraudulent activity.

[1542] Step 9:

[1543] The server comprehensively evaluates the fraud risk based on the analysis results and emotional information. For example, if the fraud rate is 80% and the user's emotional state is negative,

[1544] Step 10:

[1545] The server generates a warning message based on the fraud risk assessment results and emotional information. For example, it creates a message saying, "This person is likely to be a fraudster. Avoid lending money to them and check for more information."

[1546] Step 11:

[1547] The server sends the generated warning message to the user's device. When sending data, an encryption protocol is again used to ensure the security of the communication.

[1548] Step 12:

[1549] The user's device receives the warning message sent from the server, and the user checks the warning message on their smartphone or computer.

[1550] Step 13:

[1551] The user provides feedback on the results using a feedback form provided by the system, for example, by entering opinions such as "This analysis result is correct" or "It was misleading."

[1552] Step 14:

[1553] The user's device sends feedback to the server, which stores the received feedback and uses it to improve the AI ​​model and emotion engine in the future.

[1554] This is the specific processing flow of this system. As a result, users can detect fraudulent behavior early through their interactions on social media or dating apps, as well as recognize their own emotional state and take appropriate action.

[1555] Example 2

[1556] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1557] In conventional social networking services and dating applications, there was no system that effectively assessed and warned users about their risk of being scammed. As a result, users were more likely to fall prey to scammers' sophisticated tactics, increasing the risk of financial loss. In addition, the accuracy of warnings was low because the system did not take into account the user's emotional state.

[1558] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to input past interactions on a social networking service or a matching application; a means for transmitting the input data; a means for the processing device to input data received from the user to an AI analysis system; a means for the AI ​​analysis system to analyze the data and extract elements based on fraud patterns; a means for the processing device to extract fraud patterns from the data management system based on the analysis results; a means for the processing device to evaluate fraud risk based on the analysis results and generate a warning message; a means for notifying the user of the generated warning message; and a means for analyzing emotions in the input data and adding emotion-based information to the fraud risk evaluation. This allows users to quickly and accurately evaluate the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[1559] "User" refers to an individual who uses a social networking service or dating application.

[1560] "Social networking service" refers to an online platform that enables users to communicate with other individuals.

[1561] "Matching Application" means a mobile or web application that enables users to meet and connect with other individuals.

[1562] "Past interactions" refers to the history of messages exchanged by a user on social networking services or matching applications.

[1563] "Means for input" refers to an interface or device that allows a user to send past interactions to the system in text form.

[1564] "Processing device" refers to a central computer system for receiving, analyzing, and evaluating input data.

[1565] "Artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze data and detect patterns of fraudulent activity.

[1566] A "data management system" refers to an information storage system that stores past fraud cases and patterns and references them as needed.

[1567] "Fraud" refers to any dishonest activity that attempts to deceive users into giving them money or personal information.

[1568] "Elements" refer to the characteristics extracted by artificial intelligence analysis systems to identify patterns of fraudulent activity.

[1569] "Analysis results" refers to the conclusions and data derived by the AI ​​analysis system through data analysis.

[1570] "Fraud risk" refers to the degree of likelihood that the input data represents fraudulent activity.

[1571] "Warning Message" refers to a message containing a warning provided to a user based on a fraud risk assessment.

[1572] "Means of notification" refers to the communication means or interface used to convey a warning message to the user.

[1573] "Means for analyzing emotions" refers to systems or algorithms for extracting emotional information from text data and analyzing that information.

[1574] "Emotion-based information" refers to data that represents a user's emotional state and is used in fraud risk assessment.

[1575] This invention provides a system that inputs past interactions a user has had on social networking services or matching applications, analyzes the data using an AI analysis system and a sentiment analysis engine, and assesses the risk of fraud. This system is comprised of a user terminal, a server, an AI analysis system, a sentiment analysis engine, and a database.

[1576] Hardware and software used

[1577] User device: The smartphone or computer that the user uses to enter data and receive results. Examples: iPhone, Android device, Windows PC, Mac.

[1578] Server: A high-performance computer system to receive data and perform analytical processing. Examples: AWS EC2, Google Cloud Compute Engine.

[1579] Communication Protocol: The secure communication method used to send and receive data. For example: SSL / TLS.

[1580] Artificial intelligence analysis systems: Use machine learning algorithms that learn fraud patterns. Examples: Scikit-learn, TensorFlow.

[1581] Sentiment analysis engine: uses algorithms that analyze user sentiment. Example: IBM Watson Natural Language Understanding.

[1582] Database: A system for storing information on past fraud cases and patterns. Examples: MySQL, PostgreSQL.

[1583] System Overview

[1584] 1. User enters data:

[1585] The user inputs past interactions in text format, for example, "Hello! Would you like to meet sometime? Can you lend me a little money first?"

[1586] 2. Data transmission:

[1587] The user terminal sends the entered data to the server using an encryption protocol (such as SSL / TLS).

[1588] 3. Data Receipt and Analysis:

[1589] The server receives the data and first analyzes it with an artificial intelligence analysis system, which uses machine learning algorithms that have learned fraudulent patterns to extract elements, such as phrases like "lend me some money."

[1590] 4. Emotion analysis:

[1591] The data is passed to a sentiment analysis engine to analyze the user's emotional state. The text is classified as positive, negative, or neutral. For example, if the user's message is offensive, it will be classified as negative.

[1592] 5. Fraud risk assessment and warning message generation:

[1593] The server integrates the results of the AI ​​analysis system and the sentiment analysis engine, and compares them with past fraud patterns in the database to assess the fraud risk. For example, if the fraud risk is determined to be high (80%), a warning message is generated based on that risk.

[1594] 6. Warning message notification:

[1595] The server generates a warning message and sends it to the user's device, which displays the message "This person is likely a fraud. Avoid lending money to them and check for more information."

[1596] 7. Gathering Feedback:

[1597] Users provide feedback on the analysis results, which is sent to the server and used to improve the artificial intelligence model and sentiment analysis engine.

[1598] Specific examples

[1599] Example 1: Social media interactions

[1600] 1. Users copy social media conversations and enter them into the system.

[1601] For example: "Hi! Would you like to meet up sometime? Can I borrow a little money first?"

[1602] 2. The user terminal sends this text to the server.

[1603] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[1604] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as negative.

[1605] 5. The server calculates the probability of fraud as 80% based on the analysis results and database information.

[1606] 6. The server creates a warning message stating, "This person is likely a fraud. Avoid lending money and check for more information." and sends it to the user's device.

[1607] Example 2: Interactions on a dating app

[1608] 1. The user copies the conversations they have had on the dating app and enters them into the system.

[1609] Example: "I'd love to get married soon! I have an investment opportunity. Would you be interested?"

[1610] 2. The user terminal sends this text to the server.

[1611] 3. The server passes the received data to an artificial intelligence analysis system to extract elements that indicate fraud.

[1612] 4. At the same time, the server passes the received data to the emotion engine and classifies the user's emotion as neutral.

[1613] 5. The server calculates the probability of fraud as 90% based on the analysis results and database information.

[1614] 6. The server creates a warning message stating, "This person is very likely a scammer. Be especially careful with investment proposals and check the detailed information." and sends it to the user's terminal.

[1615] This allows users to quickly and accurately assess the risk of fraud and receive appropriate warnings. Furthermore, incorporating emotional information improves the accuracy of warnings and enables more effective responses.

[1616] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1617] Step 1:

[1618] The user inputs past interactions in text format on social networking services or dating applications. Specifically, the user inputs past messages by copying and pasting them into the application screen on their smartphone or computer. An example of input is the message "Hello! Would you like to meet sometime? Can you lend me a little money first?" The input data is saved in text format.

[1619] Step 2:

[1620] The user terminal sends the entered text data to the server. At this time, an encryption protocol such as SSL / TLS is used to ensure the security of the communication. The input is text data, and the output is encrypted data. In concrete terms, the application on the user terminal issues a send request to the server, and a progress bar is displayed.

[1621] Step 3:

[1622] The server passes the received data to an artificial intelligence analysis system, which extracts elements that indicate fraud. The data input is encrypted text data, and the output contains elements related to fraud patterns (e.g., keywords such as "money" and "lend me"). This processing uses machine learning algorithms (e.g., Scikit-learn, TensorFlow). Specifically, the server records a "receiving data" log and displays an analysis progress bar in percentage.

[1623] Step 4:

[1624] At the same time, the server passes the received data to a sentiment analysis engine to analyze the user's sentiment. The data input is the original text data, and the output is the sentiment classification result (positive, negative, neutral). A sentiment analysis algorithm (e.g., IBM Watson Natural Language Understanding) is used for sentiment analysis. Specifically, the server records a log of "sentiment analysis in progress" and displays the progress of sentiment analysis in percentage.

[1625] Step 5:

[1626] The server integrates the results of the AI ​​analysis system and sentiment analysis, and compares them with past fraud patterns in the database to assess the fraud risk. The input is the elements of the fraud pattern and the sentiment classification results, and the output is the fraud risk assessment result (e.g., fraud risk 80%). The server generates a warning message based on the analysis results. Specifically, the server records a log of "fraud risk assessment in progress" and displays the fraud risk value.

[1627] Step 6:

[1628] The server generates a warning message and sends it to the user's device. The input is the fraud risk assessment result, and the output is a warning message (e.g., "This person is likely to be a fraudster. Avoid lending money and check for more information."). Specifically, the server issues a message sending request to the user's device, and a notification is displayed on the user's device screen.

[1629] Step 7:

[1630] The user provides feedback on the system's analysis results. The input is the user's feedback (e.g., "This analysis result is accurate"), and the output is the feedback data recorded on the server. Specifically, the user enters a comment using the feedback form and clicks the submit button. The server receives the feedback and records it in a log.

[1631] The above steps enable highly accurate fraud risk assessment and the generation of warning messages, allowing users to safely use social networking sites and dating applications.

[1632] (Application example 2)

[1633] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1634] In modern society, fraudulent activities via social networking sites and other information exchange platforms are on the rise, potentially causing financial losses to many people. Furthermore, fraudulent methods are becoming increasingly sophisticated, making it difficult for users to assess the risks involved in their own communications. Therefore, there is a need for a system that can detect these fraudulent activities early and warn users. The purpose of this invention is to solve these problems and ensure user safety.

[1635] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing the user's emotions and adding an emotional evaluation to the warning message, means for notifying the user of the generated warning message, and means for generating a prompt sentence using a framework to evaluate the probability of fraud and the user's emotional state. This makes it possible to accurately evaluate the fraud risk and provide an appropriate warning that takes the user's emotional state into account in real time.

[1636] "User" refers to an entity (generally a human being) that uses a means of information exchange.

[1637] "Information exchange means" refers to systems that include software and hardware that allow users to communicate, such as social networking sites and communication apps.

[1638] "Central Computing Unit" means the central computer system that receives, analyzes, and stores data.

[1639] An "artificial intelligence analysis system" refers to a system that uses machine learning algorithms to analyze input data and extract specific patterns or elements.

[1640] A "database" refers to an information management system that references and stores past data and specific patterns based on analysis results.

[1641] "Fraud" refers to any dishonest act that involves deceiving others to obtain money or information.

[1642] "Emotion analysis" refers to the technology of determining a user's emotional state from data such as text and voice.

[1643] A "prompt" is an instruction given to a generative AI model, a piece of text that provides context for a specific analysis or generation task.

[1644] An "encryption protocol" is a technology for ensuring the security of communications when transmitting data, encrypting the data to protect it from unauthorized access by third parties.

[1645] "Warning Message" means a message intended to inform users of fraud risks or other important information.

[1646] The system of the present invention analyzes past interactions of users of social networking sites and information exchange means, assesses fraud risk and emotional state, and issues warnings. This system is composed of a user terminal, a central computing device, an AI analysis system, a database, an emotion analysis engine, and a prompt sentence generation means.

[1647] System Overview

[1648] 1. User Device

[1649] Users can use their smartphones or computers to input past interactions on social media or communication apps. For example, they can copy conversations on a chat app and input them into the system.

[1650] 2. Central computing device

[1651] This is the central computer system that receives data and runs the AI ​​analysis system and sentiment analysis engine. It also manages the database and notifies results.

[1652] 3. Artificial Intelligence Analysis System

[1653] It uses machine learning algorithms to study past fraud cases and extract elements based on fraud patterns with high accuracy, identifying phrases and patterns that indicate fraudulent activity and assessing fraud risk.

[1654] 4. Sentiment Analysis Engine

[1655] Analyzes sentiment from user input data and classifies it into positive, negative, and neutral categories, providing additional information for fraud risk assessment.

[1656] 5. Database

[1657] It is an information collection system that stores patterns and past cases of fraudulent activity. The central computing device refers to this database to assess fraud risk.

[1658] 6. Prompt generation method

[1659] The framework generates prompts to assess the probability of fraud and the emotional state of the user. The generated prompts are then fed into an AI model for specific analysis and evaluation.

[1660] Program processing explanation

[1661] The central computing unit passes the data entered by the user to the AI ​​analysis system, which uses machine learning algorithms to analyze the data and assess the risk of fraud. At the same time, the emotion analysis engine analyzes the user's emotions and classifies them as positive, negative, or neutral. Based on the analysis results and the emotion assessment, the central computing unit generates a warning message and notifies the user's device. It also generates a prompt sentence including the emotion assessment and inputs it into the AI ​​model to improve the accuracy of the analysis.

[1662] Specific examples

[1663] Example 1: Social media interactions

[1664] A user copies a conversation on social media and inputs it into the system. The content is, "Hello! Would you like to meet sometime? Can you lend me a little money first?" The central computing device passes this text to the artificial intelligence analysis system, which extracts elements that indicate fraud. At the same time, the emotion analysis engine analyzes the user's emotional state and classifies it as negative. Finally, the central computing device sends a warning message to the user's device saying, "This person is likely to be a fraudster. Be careful not to lend them money."

[1665] Prompt Sentence Examples

[1666] "What is the fraud risk of this message? We use a fraud risk assessment model."

[1667] "What is the emotional state of this message? We use sentiment analysis models."

[1668] In this way, users can analyze past interactions, understand fraud risk and emotional state, and make risk assessments and take appropriate measures.

[1669] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1670] Step 1:

[1671] A user copies past interactions on social networking sites or other information exchange platforms and inputs them into their device. The input data is in text format. This text becomes input data that is later sent from the device to the central computing device for analysis.

[1672] Step 2:

[1673] The terminal transmits the input data to the central computing device, where it is encrypted using a secure communication protocol (e.g., SSL) to protect it from unauthorized access by third parties. The output is the encrypted text data received at the central computing device.

[1674] Step 3:

[1675] The central computing unit decodes the received data and passes it to the AI ​​analysis system. The input is the decoded text data, which is preprocessed for analysis. Preprocessing includes normalizing the data and removing unnecessary characters. The output is data converted into a form suitable for analysis.

[1676] Step 4:

[1677] An AI analysis system analyzes text data and extracts elements that indicate fraud. It uses a model trained on past fraud data to assess the fraud risk from the input data. The output is a fraud risk assessment result, specifically a fraud likelihood score and a list of related elements.

[1678] Step 5:

[1679] The central computing device passes the fraud risk assessment result to the emotion analysis engine, which analyzes the user's emotions. The input is the fraud risk assessment result and text data, which the emotion analysis engine analyzes to determine the emotional state (positive, negative, neutral). The output is the emotion assessment result.

[1680] Step 6:

[1681] The central computing device integrates the fraud risk assessment results and the emotion assessment results to generate a warning message. The input is the fraud risk score and the emotion state information, and based on these, it generates an appropriate warning message for the user. The output is the warning message.

[1682] Step 7:

[1683] The generated warning message is sent to the user terminal. The central computing device sends the warning message to the user terminal, and the user receives the warning message so that the user can take appropriate action regarding the fraud risk. The output is the warning message received by the user.

[1684] Step 8:

[1685] The central computing device uses the framework to generate prompt sentences for assessing the probability of fraud and the emotional state. The inputs are the fraud risk assessment results and the emotional assessment results, and specific prompt sentences are generated based on these. The output is the prompt sentence.

[1686] Step 9:

[1687] The generated prompt sentences are input into an artificial intelligence model to improve analysis accuracy. Based on the prompt sentences, additional analysis or model retraining is performed as needed. The output is an improved analysis model or new analysis results.

[1688] In this way, at each step, the terminal, central computing device, artificial intelligence analysis system, and emotion analysis engine work together to assess fraud risk and emotional state and provide appropriate warnings.

[1689] 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.

[1690] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1691] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1692] 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.

[1693] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1694] 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.

[1695] 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).

[1696] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1697] 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."

[1698] 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.

[1699] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1700] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1701] 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.

[1702] 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.

[1703] 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.

[1704] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1705] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1706] 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.

[1707] 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.

[1708] 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.

[1709] 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.

[1710] The following is further disclosed regarding the above embodiment.

[1711] (Claim 1)

[1712] A way for users to input past interactions on social media and dating apps,

[1713] means for transmitting the input data to a server;

[1714] A means for inputting the data received by the server from the user into the AI ​​analysis system;

[1715] An AI analysis system analyzes the data and extracts elements based on fraudster patterns;

[1716] The server extracts the fraudster's pattern from the database based on the analysis results, and

[1717] a means for the server to assess fraud risk based on the analysis result and generate a warning message;

[1718] A means to notify the user of any generated warning messages.

[1719] A system including:

[1720] (Claim 2)

[1721] 10. The system of claim 1, wherein the AI ​​analytics system further comprises means for receiving user feedback and updating the model.

[1722] (Claim 3)

[1723] 10. The system of claim 1, including means for transmitting the data to be analyzed using an encryption protocol such as SSL.

[1724] "Example 1"

[1725] (Claim 1)

[1726] a means for a user to input past interactions with a communication application;

[1727] means for transmitting input data to an information processing device;

[1728] A means for inputting data received by an information processing device from a user into a machine learning model;

[1729] A machine learning model analyzes the data to extract pattern-based elements of fraudulent activity; and

[1730] A means for extracting fraudulent activity patterns from a database based on the analysis results by the information processing device;

[1731] a means for the information processing device to evaluate a fraud risk based on the analysis result and generate a warning message;

[1732] A means to notify the user of any generated warning messages.

[1733] A system including:

[1734] (Claim 2)

[1735] 10. The system of claim 1, wherein the machine learning model further comprises means for receiving user feedback and updating the model.

[1736] (Claim 3)

[1737] 10. The system of claim 1, further comprising means for transmitting the data to be analyzed using an encryption protocol.

[1738] "Application Example 1"

[1739] (Claim 1)

[1740] a means for a user to input past interactions on the communications platform;

[1741] means for transmitting input data to an information processing device;

[1742] means for inputting data received by the information processing device from a user into a knowledge base analysis system;

[1743] a knowledge-based analysis system for analyzing the data to extract pattern-based elements of fraudulent activity;

[1744] A means for extracting fraud patterns from the information accumulation system based on the analysis results by the information processing device;

[1745] a means for an information processing device to evaluate a fraud risk based on the analysis result and generate a warning message;

[1746] means for notifying a user of the generated warning text;

[1747] a mobile terminal having a user interface for copying and pasting interactions into an application;

[1748] a data transmission module having a protocol for encrypting and transmitting input data;

[1749] A system including:

[1750] (Claim 2)

[1751] 10. The system of claim 1, further comprising a knowledge-based analysis system that receives feedback via the user interface and updates the analytical model.

[1752] (Claim 3)

[1753] 10. The system of claim 1, further comprising means for transmitting the data to be analyzed using an encrypted communications protocol.

[1754] "Example 2: Combining Emotion Engines"

[1755] (Claim 1)

[1756] a means for users to input their past interactions with social networking services or dating applications;

[1757] means for transmitting input data to a processing device;

[1758] means for inputting data received by the processing device from a user into an artificial intelligence analysis system;

[1759] an artificial intelligence analysis system that analyzes the data to extract pattern-based elements of fraudulent activity;

[1760] a means for the processing device to extract fraud patterns from the data management system based on the analysis results;

[1761] means for the processing device to assess fraud risk based on the analysis result and generate a warning message;

[1762] means for notifying a user of the generated warning message;

[1763] Analyzing the sentiment of input data and adding sentiment-based information to fraud risk assessments

[1764] A system including:

[1765] (Claim 2)

[1766] 10. The system of claim 1, wherein the artificial intelligence analysis system includes means for receiving user feedback and updating the model.

[1767] (Claim 3)

[1768] 10. The system of claim 1, further comprising means for transmitting the data to be analyzed using an encryption protocol.

[1769] "Application example 2 when combining emotion engines"

[1770] (Claim 1)

[1771] a means for a user to input past interactions with the information exchange means;

[1772] means for transmitting the input data to a central computing device;

[1773] means for inputting data received by the central computing device from a user into an artificial intelligence analysis s...

Claims

1. A way for users to input past interactions on social media and dating apps, means for transmitting the input data to a server; A means for inputting the data received by the server from the user into the AI ​​analysis system; An AI analysis system analyzes the data and extracts elements based on fraudster patterns; The server extracts the fraudster's pattern from the database based on the analysis results, and a means for the server to assess fraud risk based on the analysis result and generate a warning message; A means to notify the user of any generated warning messages. A system including:

2. 10. The system of claim 1, wherein the AI ​​analysis system further comprises means for receiving user feedback and updating the model.

3. 2. The system of claim 1, further comprising means for transmitting the data to be analyzed using an encryption protocol such as SSL.

Images