System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving transaction data; means for monitoring the received transaction data in real time; means for analyzing the monitored transaction data and detecting abnormal transaction patterns; means for suspending detected abnormal transactions; means for sending the suspended transactions to a human reviewer; means for performing final transaction processing based on the reviewer's judgment; and means for notifying the user of the final processing result.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] As digital payment services become more widespread, users are increasingly facing the risk of fraudulent remittances. Such risks may cause users to feel uneasy using the services. Therefore, there is a need to improve user trust by detecting fraudulent remittances in real time and responding quickly. Specifically, technology is needed that can quickly detect abnormal transaction patterns and accurately address them. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems with a system that includes a means for receiving transaction data, a means for monitoring the received transaction data in real time, a means for analyzing the monitored transaction data to detect abnormal transaction patterns, a means for temporarily suspending the detected abnormal transactions, a means for sending the suspended transactions to a human reviewer, a means for finalizing the transaction processing based on the reviewer's judgment, and a means for notifying the user of the final processing results. By comparing the suspended transaction data with past transaction data and using an algorithm to detect abnormal patterns, abnormal transactions can be accurately identified and appropriately addressed. Furthermore, by providing a means for verifying a user's past transaction history, the recipient account, and the time period of the transaction, abnormal transactions can be prevented from being overlooked, reducing the risk of fraudulent remittances.

[0006] "Transaction Data" means data relating to the sending or receiving of money by a User using a Digital Payment Service.

[0007] "Real-time monitoring" means checking transactions as they occur immediately and monitoring them without delay.

[0008] "Abnormal transaction patterns" refer to transactions that are unnatural compared to normal transactions, including, for example, a sudden increase in transaction amount or an abnormally frequent change in remittance destination.

[0009] "Temporary suspension" refers to the process of temporarily halting a transaction when an abnormal transaction is detected and not allowing it to proceed immediately.

[0010] A "human reviewer" is a human who manually checks abnormal transactions detected by the AI ​​system and makes a final decision.

[0011] "Final Transaction Processing" means the final action taken at the stage where a transaction is approved or rejected.

[0012] "Notifying the user" refers to sending a message to inform the user that a transaction is pending or the final transaction result. [Brief explanation of the drawings]

[0013] [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

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] 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).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] This invention is a system for preventing fraudulent remittances in digital payment services. It uses AI to monitor transaction data in real time and detect abnormal transaction patterns. Furthermore, if a suspicious transaction is detected, the system temporarily suspends the transaction and leaves the final decision to a human reviewer, making a more accurate judgment.

[0035] A natural language description of what the program does

[0036] 1. Collection of transaction data

[0037] A user makes a remittance request from a terminal. For example, if a user attempts to remit 100,000 yen, transaction data (remittance destination information and transaction amount) is generated.

[0038] 2. Transmission of transaction data

[0039] The terminal transmits the generated transaction data to the server.

[0040] 3. Receiving Transaction Data

[0041] The server receives the transaction data sent from the terminal.

[0042] 4. Real-time monitoring and analysis

[0043] The server inputs the received transaction data into an AI algorithm, which analyzes the transaction data and detects abnormal patterns. For example, if a user who usually makes transactions of less than 10,000 yen suddenly transfers 100,000 yen, this would be an abnormal pattern.

[0044] 5. Detecting Abnormal Patterns

[0045] The AI ​​algorithm evaluates trading patterns against a user's past trading history and marks the transaction as suspicious if an abnormal trading pattern is detected.

[0046] 6. Suspicious Transaction Suspension

[0047] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[0048] 7. Send to human reviewers

[0049] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[0050] 8. Transaction Verification and Final Confirmation

[0051] Human reviewers check the recipient account, time of transaction, amount of transfer, etc. to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[0052] 9. Implementation of the Final Decision

[0053] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0054] To give a specific example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server and the AI ​​determines that it is an anomaly. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines that there are no problems, the transaction is finally completed. If there is a problem, the transaction is canceled and the user is notified.

[0055] This system effectively prevents fraudulent transfers via digital payment services, providing users with a service they can use with peace of mind.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[0059] Step 2:

[0060] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[0061] Step 3:

[0062] The server receives the transaction data sent from the terminal and stores it in a database.

[0063] Step 4:

[0064] The server passes the transaction data to an AI system for real-time monitoring and analysis, which uses algorithms to analyze the transaction data and detect abnormal transaction patterns.

[0065] Step 5:

[0066] The AI ​​system compares transaction data with past transaction history and statistical data to assess whether there are any unusual patterns, for example detecting unusually high transfer amounts compared to the average of a user's past transactions.

[0067] Step 6:

[0068] If the AI ​​system detects an abnormal transaction pattern, the server marks the transaction as suspicious and puts it on hold, at which point the transaction's status is set to "pending."

[0069] Step 7:

[0070] The server sends a notification to the user's terminal that "The transaction is currently pending."

[0071] Step 8:

[0072] The server sends the pending transaction data to a human reviewer, who is provided with transaction details (destination account, transaction amount, user ID, etc.).

[0073] Step 9:

[0074] Human reviewers scrutinize transaction details, such as checking the trustworthiness of the recipient and whether similar transactions have been made before.

[0075] Step 10:

[0076] If the reviewer approves the transaction, the approval result is sent to the server, and the transaction status is updated to "Approved."

[0077] Step 11:

[0078] The server resumes the approved transaction and proceeds with the actual transfer. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[0079] Step 12:

[0080] If the reviewer decides to reject the transaction, they send the rejection result to the server, and the transaction status is set to "Cancelled" because the transaction was rejected.

[0081] Step 13:

[0082] The server cancels the transaction and sends a "transaction canceled" notification to the user's terminal.

[0083] This allows the system to detect abnormal transactions in real time and take appropriate action to prevent fraudulent transfers.

[0084] Example 1

[0085] 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."

[0086] Conventional digital payment systems have difficulty detecting sophisticated fraudulent remittances, preventing users from conducting transactions with peace of mind. Furthermore, they lacked mechanisms for detecting abnormal transaction patterns in real time and taking appropriate countermeasures. This increased the risk of fraudulent transactions, potentially resulting in significant losses for financial institutions and users. This invention aims to solve these issues by utilizing AI for advanced monitoring and analysis, effectively preventing fraudulent transactions.

[0087] 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.

[0088] In this invention, the server includes means for receiving transaction data sent by a user from a terminal, means for transmitting the received transaction data to the server via a secure communication means, means for storing the transaction data received by the server within the system, means for inputting the received transaction data into a generative AI model, means for monitoring and analyzing the transaction data in real time using the generative AI model, means for detecting anomalous transaction patterns from the analyzed transaction data, means for temporarily suspending the detected anomalous transaction and setting its status to "suspended," means for notifying the user of the suspended transaction data, means for transmitting the suspended transaction to a human reviewer, means for the reviewer to carefully examine the transaction content and make a judgment, means for final transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This enables real-time monitoring of transaction data and early detection of anomalies, thereby significantly reducing the risk of fraudulent transactions.

[0089] "Transaction data" is a collection of information related to remittances and payments made by users, including details such as the recipient account information and the remittance amount.

[0090] "Terminal" refers to the digital device used by a user, such as a smartphone or tablet, that provides the interface for making a remittance request.

[0091] A "server" is a computer system that receives, processes, and analyzes transaction data sent from a terminal.

[0092] A "generative AI model" is a model that uses artificial intelligence to analyze trading data and includes algorithms to detect anomalous trading patterns.

[0093] "Real-time monitoring" is the process of monitoring transaction data in real time and analyzing it immediately.

[0094] "Abnormal transaction patterns" refer to potentially fraudulent transaction behavior that deviates from normal transaction patterns and is detected by the AI ​​model.

[0095] "Suspicious transactions" are transactions that are marked as anomalous by the AI ​​model and require further review.

[0096] "Suspension" refers to a state in which a transaction in which an abnormality has been detected is temporarily halted and a final decision is pending.

[0097] A "reviewer" is a professional who examines pending transactions from a human perspective and makes a final decision.

[0098] "Final transaction processing" refers to the series of processes that approve or reject a transaction based on the reviewer's judgment.

[0099] "Notifications" are messages that inform users of the status or final outcome of a transaction.

[0100] The above definitions allow understanding of important terms related to this invention.

[0101] This invention is a system for preventing fraudulent remittances in digital payment services, specifically a system for real-time monitoring of transaction data and detection of abnormal patterns. The system includes major components such as users, terminals, and servers.

[0102] First, a user makes a remittance request using a device such as a smartphone or tablet. The user enters the recipient account information and the amount to be remitted, and transaction data is generated on the device. The generated transaction data includes the recipient information and the transaction amount.

[0103] The terminal encrypts the generated transaction data and then sends it to the server using a secure communication method (e.g., HTTPS). The server stores the transaction data received from the terminal in its system and verifies that the data is complete and accurate.

[0104] The server then inputs the received transaction data into a generative AI model, an algorithm that analyzes transaction data in real time against historical transaction data to detect anomalous transaction patterns. The AI ​​model identifies transactions that deviate from normal transaction patterns and marks transactions that are deemed potentially fraudulent as suspicious.

[0105] Any abnormal transaction detected will be temporarily suspended by the server and its status will be set to "Pending." At the same time, the server will send a notification to the user saying "The transaction is currently pending." This will let the user know that the transaction is pending and undergoing final confirmation.

[0106] The server then sends the pending transaction data to a human reviewer who examines the transaction details, checking the recipient account, time of day, amount, etc., to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[0107] If the reviewer approves the transaction, the server will re-promote the transaction, or if the transaction is rejected, it will cancel the transaction and send a "Transaction Cancelled" notification to the user.

[0108] For example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, where it is identified as an anomaly by the AI ​​model. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines there are no problems, the transaction is finally completed. If there are problems, the transaction is canceled and the user is notified.

[0109] An example of an input prompt for a generative AI model is:

[0110] Analyze the transaction data below to determine if there is any possibility of fraud.

[0111] Transaction Data:

[0112] User ID: 12345

[0113] Remittance destination: ABC Bank

[0114] Amount to be sent: 100,000 yen

[0115] Past trading patterns: Over 90% of transactions were under 10,000 yen

[0116] This transaction is unusual because it deviates significantly from the normal pattern.

[0117] This system effectively prevents fraudulent remittances in digital payment services, providing users with a service they can use with peace of mind. Also, through specific operational procedures, it is possible to understand in detail how this invention is implemented.

[0118] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0119] Step 1: Collect transaction data

[0120] The user makes a remittance request using the terminal. The user enters the remittance account information and the remittance amount, and transaction data is generated. This transaction data includes the remittance account information and the transaction amount.

[0121] (Input) The remittance destination information and transaction amount entered by the user into the terminal.

[0122] (Output) The generated transaction data.

[0123] Specific example of operation: The user opens the app, enters the recipient account information and the amount to send, and presses the "Send" button to generate transaction data.

[0124] Step 2: Sending transaction data

[0125] The terminal sends the generated transaction data to the server, which transmits the data using a secure communication method (e.g. HTTPS).

[0126] (Input) Transaction data generated by the terminal.

[0127] (Output) The encrypted transaction data sent to the server.

[0128] Specific operation example: The transaction data generated by the terminal is encrypted and sent to the server using the REST API.

[0129] Step 3: Receiving transaction data

[0130] The server receives the transaction data sent from the terminal, verifies that the data is complete and accurate, and stores the data in the system.

[0131] (Input) Encrypted transaction data sent from the terminal.

[0132] (Output) Transaction data stored in the system.

[0133] Specific operation example: The server saves the received data in a database and returns a reception confirmation response to the terminal.

[0134] Step 4: Real-time monitoring and analysis

[0135] The server inputs the received transaction data into a generative AI model, which analyzes the transaction data and detects anomalous patterns.

[0136] (Input) Transaction data stored on the server.

[0137] (Output) Trade valuation analyzed by the AI ​​model.

[0138] Specific example of operation: The server inputs transaction data into the generative AI model, and the AI ​​compares it with past transaction data and performs analysis.

[0139] Step 5: Detecting abnormal patterns

[0140] The generative AI model evaluates trading patterns against historical transaction data and marks transactions as suspicious if an abnormal trading pattern is detected.

[0141] (Input) Trade valuation analyzed by the AI ​​model.

[0142] (Output) Transactions marked as suspicious.

[0143] Specific example of operation: Transactions that the AI ​​determines to be abnormal are marked as "suspicious" and information about the suspicious transaction is recorded in a log.

[0144] Step 6: Suspicious Transaction Suspension

[0145] The server temporarily holds the suspicious transaction and sets its status to "Pending." The user is notified that "The transaction is currently pending."

[0146] (Input) Transactions marked as suspicious.

[0147] (Output) Transaction status set to "Pending" and user notification.

[0148] Example of specific behavior: The server updates the transaction status to "pending" and sends a pending notification to the user.

[0149] Step 7: Send to human reviewers

[0150] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[0151] (Input) "Pending" transaction data.

[0152] (Output) The transaction data sent to the reviewer.

[0153] Specific operation example: The server sends pending transaction data to the reviewer's system, and the reviewer receives the transaction data and starts reviewing it.

[0154] Step 8: Transaction verification and final confirmation

[0155] Human reviewers check the destination account, time of transaction, amount sent, etc. to determine whether the transaction is legitimate.

[0156] (Input) The transaction data received by the reviewer.

[0157] (Output) The transaction verification result.

[0158] Example of specific operation: A reviewer checks the remittance destination, transaction amount, etc., and asks the user for additional confirmation if necessary.

[0159] Step 9: Making a final decision

[0160] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0161] (Input) Transaction verification result.

[0162] (Output) Final processing results and user notification.

[0163] Example of specific operation: A reviewer makes a decision on the system to approve or reject a transaction, and the server proceeds with approved transactions, cancels rejected transactions, and sends a final notification to the user.

[0164] (Application example 1)

[0165] 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."

[0166] The lack of a system to effectively prevent fraudulent transfers is a problem for existing digital payment services. In particular, it is difficult to monitor transaction data in real time, detect anomalies, temporarily suspend suspicious transactions, and promptly notify users and reviewers. This can lead to a lack of swift action when fraudulent transactions occur, which can result in a loss of user trust.

[0167] 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.

[0168] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for analyzing the monitored transaction data and detecting abnormal transaction patterns, means for temporarily suspending the detected abnormal transactions, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, means for notifying the user of the final processing result, and means for automatically notifying the user and the reviewer by email when an abnormal transaction is detected. This makes it possible to detect abnormal transactions in real time and respond to them quickly and efficiently.

[0169] "Transaction data" refers to information relating to transfers and payments made by users, including details such as the transfer destination, transfer amount, time, and transaction ID.

[0170] "Real-time monitoring means" refers to methods and systems for capturing transaction data in real time and analyzing it without delay.

[0171] "Means for detecting abnormal trading patterns" refers to AI algorithms or rule-based systems that identify transactions that are unusual or risky compared to normal trading history.

[0172] "Means for temporarily suspending abnormal transactions" refers to a system that has the ability to immediately stop and suspend a suspicious transaction when it is detected.

[0173] "Means of submission to human reviewer" refers to the communications methods and protocols used to submit details of a pending transaction to a human reviewer for final review.

[0174] "Means for finalizing transaction processing based on reviewer's judgment" refers to the system functionality for executing or canceling a transaction based on the reviewer's evaluation and decision.

[0175] "Means for notifying the user of the final processing result" refers to the notification method used to inform the user of the result of a transaction after it has been completed or canceled.

[0176] "Means for automatic email notification" refers to a system function that automatically sends emails to users and reviewers based on pre-set templates when an anomaly in transaction data is detected.

[0177] The system for realizing the present invention is configured based on a program including the following processing procedures.

[0178] First, a user makes a remittance request using a terminal. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This transaction data is sent to the server in real time.

[0179] The server receives the received transaction data and begins monitoring it in real time. It uses AI algorithms such as IsolationForest to detect abnormal transaction patterns and identify transactions that are unusual or high risk compared to normal transaction history (e.g., when transactions of several thousand yen are usually made, but a large amount is suddenly transferred).

[0180] Transactions that are identified as anomalous are temporarily put on hold and the transaction status is set to "Pending." Details of the pending transaction are sent to the user and human reviewer using an automated email notification feature. The user is notified that the transaction is currently on hold, and the reviewer is provided with details of the transaction.

[0181] Next, a reviewer will examine the submitted transaction data to determine whether or not the transaction is abnormal. This review will take into account the user's past transaction history, remittance destination information, and the time of the transaction. Based on the reviewer's decision, the final transaction processing will be carried out. If the transaction is approved, it will proceed; if it is rejected, it will be canceled.

[0182] The final transaction processing result will be notified to the user. Notifications will include messages such as "Transaction completed" or "Transaction canceled" using an automated email notification function.

[0183] This system mainly uses the following hardware and software:

[0184] Hardware: Devices such as smartphones

[0185] Software: Python, pandas (data processing), scikit-learn (AI modeling), smtplib (email sending)

[0186] For example, if a user normally sends a few thousand yen, but suddenly attempts to send 150,000 yen, the AI ​​will detect this suspicious transaction and put it on hold. The user and reviewers will be notified, and the reviewers will determine the legitimacy of the transaction.

[0187] An example prompt is:

[0188] Please provide a detailed review of the following transaction.

[0189] User ID: 1

[0190] Amount to be sent: 150,000 yen

[0191] Timestamp: 2023-10-10 16:00:00

[0192] New Recipient: True

[0193] The transaction is currently pending. You have been notified and can now decide whether to accept or cancel the transaction.

[0194] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0195] Step 1:

[0196] A user makes a remittance request using a smartphone. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This generates transaction data. The inputs are the "remittance amount," "remittance destination information," and "timestamp," and the output is "transaction data."

[0197] Step 2:

[0198] The terminal sends the generated transaction data to the server. Based on the transaction data, the server receives the data. The input here is "transaction data" and the output is "transaction data received by server."

[0199] Step 3:

[0200] The server begins monitoring transaction data in real time. The received transaction data is input into an AI algorithm such as IsolationForest, which analyzes it to detect abnormal transaction patterns. The input is "transaction data" and the output is "detection results of abnormal patterns." Specifically, each field of the transaction data (such as the amount sent and the recency of the recipient) is applied to the AI ​​algorithm.

[0201] Step 4:

[0202] Transactions that are determined to be abnormal are temporarily suspended and the transaction status is set to "pending." This suspension process causes the server to mark the abnormal transaction as a flag and handle it specially in subsequent processes. The input is the "detection result of an abnormal transaction," and the output is the "pending transaction data."

[0203] Step 5:

[0204] The server automatically sends emails with details of the released transaction to the user and reviewer. Specifically, emails are sent using smtplib. The input is "pending transaction data" and the output is "sent email."

[0205] Step 6:

[0206] The reviewer scrutinizes the transaction data received by email and determines whether or not the transaction is anomalous. The input here is the "transaction data received by email" and the output is the "review results." Specifically, the reviewer determines the validity of the anomalous transaction based on the transaction history, remittance destination information, transaction time period, etc.

[0207] Step 7:

[0208] Based on the reviewer's decision, the server performs the final transaction processing. If the transaction is approved, the transaction data is processed as usual. If it is rejected, the transaction is canceled. The input is the "review result" and the output is the "final transaction processing result."

[0209] Step 8:

[0210] The server notifies the user of the final processing result. This notification is sent via automatic email using smtplib. The input is the "final transaction processing result" and the output is "notification to the user." Specifically, an email is sent to the user stating the status of the transaction, whether it was approved or canceled.

[0211] 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.

[0212] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[0213] A natural language description of what the program does

[0214] 1. Collection of transaction data

[0215] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[0216] 2. Transmission of transaction data

[0217] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[0218] 3. Receiving Transaction Data

[0219] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it also starts the user's emotion engine.

[0220] 4. Acquiring Emotion Data

[0221] It captures emotional data about users through cameras, microphones, and other sensors built into their devices, such as facial expressions, tone of voice, and stress levels.

[0222] 5. Emotion Data Analysis

[0223] The emotion engine analyzes the acquired emotion data and determines the user's emotional state (for example, impatience, tension, relief).

[0224] 6. Real-time monitoring and analysis

[0225] The server then passes the received transaction data to the AI ​​system for real-time monitoring and analysis, which compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns.

[0226] 7. Detecting abnormal patterns and emotional states

[0227] The AI ​​system will mark a transaction as suspicious if it detects an abnormal trading pattern, and an emotional engine will make an additional determination as to whether a transaction is anomalous based on the user's emotional state.

[0228] 8. Suspicious Transaction Suspension

[0229] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[0230] 9. Send to human reviewers

[0231] The server sends the pending transaction data and associated sentiment data to human reviewers, who then scrutinize the data.

[0232] 10. Transaction Verification and Final Confirmation

[0233] Human reviewers consider the destination account, transaction time, transaction amount, and captured emotional state to make the final decision.

[0234] 11. Implementation of Final Decision

[0235] If the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0236] To give a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they are deemed legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[0237] This system will effectively prevent fraudulent transfers and irregular transactions in digital payment services, providing users with a service they can use with greater peace of mind.

[0238] The processing flow will be explained below.

[0239] Step 1:

[0240] The user inputs a remittance request using the terminal, specifies the recipient's account information and the amount to be remitted, and presses the remit button.

[0241] Step 2:

[0242] The terminal sends the generated transaction data (recipient account information, remittance amount, user ID, transaction time) to the server.

[0243] Step 3:

[0244] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it starts the emotion engine and starts acquiring the user's emotion data.

[0245] Step 4:

[0246] The device collects emotional data from users through built-in cameras, microphones, and other sensors, such as facial expressions, tone of voice, and heart rate.

[0247] Step 5:

[0248] The device sends the collected emotion data to the server, which then passes the data to the emotion engine.

[0249] Step 6:

[0250] The emotion engine analyzes the transmitted emotion data and determines the user's emotional state (e.g., anxious, nervous, relieved).

[0251] Step 7:

[0252] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with past transaction history and statistical data to detect abnormal patterns.

[0253] Step 8:

[0254] If the AI ​​system detects an abnormal transaction pattern, for example, if an amount that is excessively high compared to the amount normally sent, it will notify the server.

[0255] Step 9:

[0256] The server combines the results of the AI ​​system and the analysis results of the emotion engine to make a final decision on whether a transaction is suspicious. For example, if a transaction is abnormal and the user is in a tense state, the server will determine that the transaction is suspicious.

[0257] Step 10:

[0258] The server will temporarily suspend the suspicious transaction, set the transaction status to "Pending," and send a notification to the user stating, "The transaction is currently pending."

[0259] Step 11:

[0260] The server sends the pending transaction data and associated emotion data to a human reviewer, who is provided with the transaction details and emotional state.

[0261] Step 12:

[0262] Human reviewers will make a final decision based on the destination account, transaction time, transaction amount, and captured emotional data, taking into account, for example, if the transaction amount is significantly larger than normal or if the user appears to be in an unnatural state of stress.

[0263] Step 13:

[0264] If the reviewer approves the transaction, they send the approval result to the server, which updates the transaction status to "approved" and allows the transaction to proceed.

[0265] Step 14:

[0266] If the transaction is approved, the server proceeds with the actual transfer process. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[0267] Step 15:

[0268] If the reviewer decides to reject the transaction, the rejection is sent to the server, which updates the transaction status to "Cancelled" and cancels the transaction.

[0269] Step 16:

[0270] If the transaction is cancelled, the server sends a notification to the user's terminal that the transaction has been cancelled.

[0271] In this way, this system can combine and analyze transaction data and emotional data to detect and prevent suspicious transactions with high accuracy.

[0272] Example 2

[0273] 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."

[0274] In recent years, fraudulent remittances and abnormal transactions in digital payment services have been increasing, and conventional security measures are often inadequate to address these issues. New technologies are needed to detect fraudulent activity in advance and respond quickly. Furthermore, taking the user's emotional state into account is expected to enable more accurate detection of abnormal transactions, but achieving this requires advanced analytical technology and real-time processing. The present invention aims to solve these problems.

[0275] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving transaction data, means for monitoring the transaction data in real time, means for analyzing the transaction data and detecting abnormal transaction patterns, means for acquiring and analyzing user emotion data to identify the user's emotional state, means for temporarily suspending abnormal transactions based on the abnormal transactions and the emotion analysis results, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing results. This makes it possible to detect fraudulent transactions with high accuracy from both the transaction data and the emotion data, and to prevent fraud quickly and reliably.

[0276] "Transaction data" refers to data that includes detailed information such as the recipient account information specified by the user, the amount to be transferred, the user ID, and the time of the transaction.

[0277] "Real-time monitoring" means that as soon as transaction data is generated, its contents are immediately investigated and processing is carried out to detect any fraud or abnormalities.

[0278] "Abnormal trading patterns" refer to trading patterns that are clearly different from normal trading behavior or past trading history, or that are suspected of being fraudulent.

[0279] "Emotional data" refers to data regarding a user's emotional state, such as a user's facial expression, tone of voice, stress level, etc.

[0280] An "emotion engine" refers to software or algorithms that analyze emotional data and identify a user's emotional state.

[0281] "Temporary suspension" means that the abnormal transaction detected will not be immediately processed, but will be temporarily stopped for further inspection and confirmation.

[0282] "Reviewer" refers to a human professional who reviews transaction data and sentiment data to make a final determination regarding suspicious transactions.

[0283] "Final transaction processing" refers to the process of proceeding with or canceling a transfer based on the reviewer's determination of whether the transaction is fraudulent or not.

[0284] "Notification" refers to messages or information that inform a user about the status of a transaction or the final processing result.

[0285] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[0286] First, the user opens the application on the terminal and initiates a transaction. They enter a transfer request and specify the recipient's account information and the amount to be transferred. When the user presses the "Transfer" button, this information is generated as transaction data by the terminal. The terminal then encrypts the generated transaction data and sends it to the server via a secure network. The transaction data includes details such as the recipient's account information, the amount to be transferred, the user ID, and the time of the transaction.

[0287] The server receives transaction data sent from the terminal in real time and stores it in a database. The server also sends commands to the user's device to activate the camera, microphone, and other sensors to obtain the user's emotional data. The device collects the user's facial expressions, tone of voice, stress level, and other information in real time and generates emotional data. This emotional data is sent back to the server and analyzed by an emotion engine to identify the user's emotional state. For example, it determines states such as "anxiety" or "tension."

[0288] The server then passes the received transaction data to an AI system for real-time monitoring and analysis. The AI ​​system compares the transaction data with past transaction patterns and statistical data to detect abnormal transaction patterns. Based on the detected abnormal transactions and the results of sentiment analysis, the system temporarily suspends suspicious transactions and sets their status to "on hold." The server then sends a notification to the user stating, "The transaction is currently on hold."

[0289] The server then sends the pending transaction data and associated emotion data to a human reviewer who reviews the data and makes a final decision. The human reviewer reviews the destination account, transaction time, transaction amount, and captured emotion state. If necessary, the human reviewer can provide additional confirmation to the user.

[0290] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0291] This system effectively prevents fraudulent transfers and irregular transactions in digital payment services, allowing users to use the service with peace of mind.

[0292] As a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​system determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they determine that it is legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[0293] Prompt Sentence Examples

[0294] "Describe a scenario where a user is attempting to transfer a large amount of money. Detail how you would process it by analyzing sentiment data and detecting anomalous transactions."

[0295] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0296] Step 1:

[0297] The user opens the application on the terminal and enters a remittance request. The data entered includes the recipient's account information, the remittance amount, the user ID, and the transaction time. When the user presses the "Remit" button, this data is generated as transaction data by the terminal. The terminal then encrypts this transaction data and sends it to the server over a secure network.

[0298] Input: Remittance account information, remittance amount, user ID, transaction time

[0299] Output: Encrypted transaction data

[0300] Step 2:

[0301] The server receives the transaction data sent from the terminal and stores it in a database. The server also starts the emotion engine and prepares the user's device to collect emotion data.

[0302] Input: Encrypted transaction data

[0303] Output: Transaction data stored in a database

[0304] Step 3:

[0305] The server sends commands to the user's device to activate the camera, microphone, and other sensors, and the device collects the user's facial expressions, tone of voice, stress level, and other information in real time to generate emotional data.

[0306] Input: Commands to the user device

[0307] Output: User emotion data

[0308] Step 4:

[0309] Emotional data is sent from the user's device to the server, which then analyzes it with an emotion engine to identify the user's emotional state. For example, it determines whether the user is feeling anxious or nervous.

[0310] Input: Emotion data

[0311] Output: Parsed emotional state

[0312] Step 5:

[0313] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with historical transaction patterns and statistical data to detect abnormal transaction patterns.

[0314] Input: Transaction data

[0315] Output: Abnormal trading pattern detection results

[0316] Step 6:

[0317] If the AI ​​system detects an abnormal trading pattern, it will mark the transaction as suspicious, and if the emotion engine determines that the transaction is suspicious based on the user's emotional state, the transaction will also be marked as suspicious.

[0318] Input: Detected abnormal trading patterns, analyzed emotional state

[0319] Output: Suspicious transaction mark

[0320] Step 7:

[0321] The server temporarily holds the suspicious transaction and sets its status to "Pending." The server sends a notification to the user saying, "The transaction is currently pending."

[0322] Input: Mark suspicious transaction

[0323] Output: Pending transaction status, notification to user

[0324] Step 8:

[0325] The server sends the pending transaction data and associated sentiment data to human reviewers, who scrutinize the data and check for anomalies.

[0326] Input: Pending transaction data, associated sentiment data

[0327] Output: Send data to reviewers

[0328] Step 9:

[0329] A human reviewer will review the details of the destination account, transaction time, transaction amount, and captured emotional state. If necessary, the reviewer will ask the user for additional confirmation.

[0330] Input: Data to reviewer

[0331] Output: Final decision

[0332] Step 10:

[0333] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0334] Input: Final decision

[0335] Output: Transaction proceeds or is cancelled, and the user is notified

[0336] (Application example 2)

[0337] 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."

[0338] Increasing fraudulent remittances and abnormal transactions are becoming a major problem in modern digital payment services. This puts users at risk of fraudulent transactions and may undermine trust in payment services. Furthermore, conventional systems detect anomalies solely by analyzing transaction data, which means there is a risk of overlooking fraudulent transactions. Furthermore, accurate judgments are difficult because they do not take into account the user's emotional state. Therefore, a more accurate fraud detection system is needed.

[0339] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0340] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for acquiring user emotional data, means for analyzing the acquired emotional data to determine the user's emotional state, means for analyzing the monitored transaction data and temporarily suspending transactions determined to be anomalous based on anomalous transaction patterns and the user's emotional state, means for transmitting the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This makes it possible to improve the accuracy of detecting anomalous transactions and effectively reduce the risk of fraudulent remittances by taking the user's emotional state into consideration in addition to analyzing the transaction data.

[0341] "Transaction data" is information generated when a user transfers or makes a payment, and includes details such as the recipient's account information, the amount transferred, the user ID, and the time of the transaction.

[0342] An "abnormal transaction pattern" is a transaction that shows unnatural trends that differ from normal transactions and is an indicator of possible fraudulent transfers or other fraudulent activity.

[0343] "Emotion data" is data that reflects the user's emotional state, obtained by analyzing the user's facial expressions, tone of voice, stress level, etc.

[0344] The "emotional state" indicates the psychological state of the user, and refers to states such as impatience, tension, and relief.

[0345] "Temporary suspension" refers to the process of temporarily halting a transaction that has been determined to be abnormal and withholding a final decision.

[0346] A "human reviewer" is someone who manually reviews transactions and sentiment data held by the system and ultimately approves or rejects the transaction.

[0347] The system that realizes this application example consists of the following steps: First, a user launches a digital payment service application using a smartphone terminal. When the user enters a remittance request and specifies the recipient's account information and the remittance amount, the smartphone terminal generates transaction data and sends it to the server.

[0348] The server stores the received transaction data in a database. The transaction data includes details such as the recipient's account information, the remittance amount, the user ID, and the transaction time. At the same time, the user's emotion engine runs on the smartphone and collects the user's emotion data through the device's camera and microphone.

[0349] Emotion data includes the user's facial expressions, tone of voice, stress level, etc. The emotion engine analyzes the emotion data in real time to determine the user's emotional state. This emotion data is then sent to the server and integrated with the transaction data.

[0350] After integrating the above data, the server's AI system uses the received transaction data and emotional state data for real-time monitoring and analysis. The AI ​​system compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns, and then takes the emotional state into account to make a final anomalous transaction judgment. Transactions that are judged to be abnormal are temporarily suspended and sent to human reviewers for a final decision.

[0351] The reviewer verifies the validity of the transaction based on the pending transaction data and the user's emotional state data sent from the server. Based on the review results, the server performs the final transaction processing and notifies the user of the result.

[0352] The specific hardware and software used are as follows:

[0353] Smartphone device: camera, microphone, processing unit

[0354] Server: Database management system (MySQL, etc.), AI model (TensorFlow / PyTorch)

[0355] Emotion analysis engine: Voice analysis library (Librosa, etc.), face recognition library (OpenCV, etc.)

[0356] Monitoring system: System monitoring tools (Prometheus, etc.)

[0357] As a concrete example, consider a situation where a user attempts to transfer more than 100,000 yen. If the AI ​​detects an abnormality during the transfer process, such as the camera detecting a panicked expression or the microphone detecting a tense tone of voice, the transaction will be temporarily put on hold. The server then sends this data to a human reviewer, who will then approve or reject the transaction based on their final judgment.

[0358] Example prompt sentence:

[0359] "If a user attempts to send more than 100,000 yen using a smartphone app, consider a system that verifies the intention of the transfer. AI will monitor abnormal transactions and the user's facial expressions in real time to determine safety."

[0360] This invention improves the accuracy of detecting abnormal transactions by combining analysis of transaction data with data on the user's emotional state, making it possible to effectively reduce the risk of fraudulent remittances.

[0361] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0362] Step 1:

[0363] A user launches a digital payment service application on a smartphone terminal. The user enters a remittance request, specifying the recipient's account information and the amount to be sent. When the "Remit" button is pressed, the terminal generates transaction data.

[0364] Input: Recipient account information, remittance amount, user ID, transaction time

[0365] Output: Transaction data

[0366] Step 2:

[0367] The terminal transmits the generated transaction data to the server.

[0368] Input: Transaction data

[0369] Output: Transaction data sent to the server

[0370] Step 3:

[0371] The server stores the received transaction data in a database.

[0372] Input: Transaction data sent from the terminal

[0373] Output: Transaction data stored in a database

[0374] Step 4:

[0375] At the same time, the device's camera and microphone are used to collect the user's emotional data: the camera captures the user's facial expressions, and the microphone captures the tone of voice and stress level.

[0376] Input: Camera video, audio data

[0377] Output: Emotion data

[0378] Step 5:

[0379] The device analyzes the collected emotional data and determines the user's emotional state. Analysis is performed using a face recognition library (such as OpenCV) and a voice analysis library (such as Librosa).

[0380] Input: Emotion data

[0381] Output: Emotional state (e.g., anxious, nervous, relieved)

[0382] Step 6:

[0383] The device transmits the analyzed emotional state to the server.

[0384] Input: Parsed emotional state

[0385] Output: Emotional state sent to the server

[0386] Step 7:

[0387] The server combines the received transaction data with the emotional state data and passes it to the AI ​​system.

[0388] Input: Transaction data, emotional state data

[0389] Output: Integrated data (transaction data + emotional state data)

[0390] Step 8:

[0391] The AI ​​system monitors and analyzes the integrated data in real time, comparing it with previous trading patterns and statistical data to detect abnormal patterns, and then takes into account emotional states to make a final judgment on abnormal trading.

[0392] Inputs: Consolidated data, historical trading patterns, statistical data

[0393] Output: Abnormal transaction determination result

[0394] Step 9:

[0395] The server temporarily holds transactions that are identified as abnormal and sends the transaction data and emotional state data to a human reviewer.

[0396] Input: Abnormal transaction judgment results, transaction data, emotional state data

[0397] Output: Pending transactions, reviewer notifications

[0398] Step 10:

[0399] The reviewer evaluates the validity of the transaction based on the pending transaction data and emotional state data. If the transaction is deemed valid, the server proceeds with the transaction, but if there are any problems, the server cancels the transaction.

[0400] Input: Pending transaction data, emotional state data

[0401] Output: Review result (approved or rejected)

[0402] Step 11:

[0403] The server performs the final transaction processing based on the review results and notifies the user of the results.

[0404] Input: Review results

[0405] Output: Notification of final transaction result (transfer completion or transaction cancellation notification)

[0406] 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.

[0407] 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.

[0408] 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.

[0409] [Second embodiment]

[0410] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0411] 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.

[0412] 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).

[0413] 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.

[0414] 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.

[0415] 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).

[0416] 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.

[0417] 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.

[0418] 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.

[0419] 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.

[0420] In the smart glasses 214, 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.

[0421] 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."

[0422] This invention is a system for preventing fraudulent remittances in digital payment services. It uses AI to monitor transaction data in real time and detect abnormal transaction patterns. Furthermore, if a suspicious transaction is detected, the system temporarily suspends the transaction and leaves the final decision to a human reviewer, making a more accurate judgment.

[0423] A natural language description of what the program does

[0424] 1. Collection of transaction data

[0425] A user makes a remittance request from a terminal. For example, if a user attempts to remit 100,000 yen, transaction data (remittance destination information and transaction amount) is generated.

[0426] 2. Transmission of transaction data

[0427] The terminal transmits the generated transaction data to the server.

[0428] 3. Receiving Transaction Data

[0429] The server receives the transaction data sent from the terminal.

[0430] 4. Real-time monitoring and analysis

[0431] The server inputs the received transaction data into an AI algorithm, which analyzes the transaction data and detects abnormal patterns. For example, if a user who usually makes transactions of less than 10,000 yen suddenly transfers 100,000 yen, this would be an abnormal pattern.

[0432] 5. Detecting Abnormal Patterns

[0433] The AI ​​algorithm evaluates trading patterns against a user's past trading history and marks the transaction as suspicious if an abnormal trading pattern is detected.

[0434] 6. Suspicious Transaction Suspension

[0435] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[0436] 7. Send to human reviewers

[0437] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[0438] 8. Transaction Verification and Final Confirmation

[0439] Human reviewers check the recipient account, time of transaction, amount of transfer, etc. to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[0440] 9. Implementation of the Final Decision

[0441] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0442] To give a specific example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server and the AI ​​determines that it is an anomaly. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines that there are no problems, the transaction is finally completed. If there is a problem, the transaction is canceled and the user is notified.

[0443] This system effectively prevents fraudulent transfers via digital payment services, providing users with a service they can use with peace of mind.

[0444] The processing flow will be explained below.

[0445] Step 1:

[0446] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[0447] Step 2:

[0448] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[0449] Step 3:

[0450] The server receives the transaction data sent from the terminal and stores it in a database.

[0451] Step 4:

[0452] The server passes the transaction data to an AI system for real-time monitoring and analysis, which uses algorithms to analyze the transaction data and detect abnormal transaction patterns.

[0453] Step 5:

[0454] The AI ​​system compares transaction data with past transaction history and statistical data to assess whether there are any unusual patterns, for example detecting unusually high transfer amounts compared to the average of a user's past transactions.

[0455] Step 6:

[0456] If the AI ​​system detects an abnormal transaction pattern, the server marks the transaction as suspicious and puts it on hold, at which point the transaction's status is set to "pending."

[0457] Step 7:

[0458] The server sends a notification to the user's terminal that "The transaction is currently pending."

[0459] Step 8:

[0460] The server sends the pending transaction data to a human reviewer, who is provided with transaction details (destination account, transaction amount, user ID, etc.).

[0461] Step 9:

[0462] Human reviewers scrutinize transaction details, such as checking the trustworthiness of the recipient and whether similar transactions have been made before.

[0463] Step 10:

[0464] If the reviewer approves the transaction, the approval result is sent to the server, and the transaction status is updated to "Approved."

[0465] Step 11:

[0466] The server resumes the approved transaction and proceeds with the actual transfer. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[0467] Step 12:

[0468] If the reviewer decides to reject the transaction, they send the rejection result to the server, and the transaction status is set to "Cancelled" because the transaction was rejected.

[0469] Step 13:

[0470] The server cancels the transaction and sends a "transaction canceled" notification to the user's terminal.

[0471] This allows the system to detect abnormal transactions in real time and take appropriate action to prevent fraudulent transfers.

[0472] Example 1

[0473] 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."

[0474] Conventional digital payment systems have difficulty detecting sophisticated fraudulent remittances, preventing users from conducting transactions with peace of mind. Furthermore, they lacked mechanisms for detecting abnormal transaction patterns in real time and taking appropriate countermeasures. This increased the risk of fraudulent transactions, potentially resulting in significant losses for financial institutions and users. This invention aims to solve these issues by utilizing AI for advanced monitoring and analysis, effectively preventing fraudulent transactions.

[0475] 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.

[0476] In this invention, the server includes means for receiving transaction data sent by a user from a terminal, means for transmitting the received transaction data to the server via a secure communication means, means for storing the transaction data received by the server within the system, means for inputting the received transaction data into a generative AI model, means for monitoring and analyzing the transaction data in real time using the generative AI model, means for detecting anomalous transaction patterns from the analyzed transaction data, means for temporarily suspending the detected anomalous transaction and setting its status to "suspended," means for notifying the user of the suspended transaction data, means for transmitting the suspended transaction to a human reviewer, means for the reviewer to carefully examine the transaction content and make a judgment, means for final transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This enables real-time monitoring of transaction data and early detection of anomalies, thereby significantly reducing the risk of fraudulent transactions.

[0477] "Transaction data" is a collection of information related to remittances and payments made by users, including details such as the recipient account information and the remittance amount.

[0478] "Terminal" refers to the digital device used by a user, such as a smartphone or tablet, that provides the interface for making a remittance request.

[0479] A "server" is a computer system that receives, processes, and analyzes transaction data sent from a terminal.

[0480] A "generative AI model" is a model that uses artificial intelligence to analyze trading data and includes algorithms to detect anomalous trading patterns.

[0481] "Real-time monitoring" is the process of monitoring transaction data in real time and analyzing it immediately.

[0482] "Abnormal transaction patterns" refer to potentially fraudulent transaction behavior that deviates from normal transaction patterns and is detected by the AI ​​model.

[0483] "Suspicious transactions" are transactions that are marked as anomalous by the AI ​​model and require further review.

[0484] "Suspension" refers to a state in which a transaction in which an abnormality has been detected is temporarily halted and a final decision is pending.

[0485] A "reviewer" is a professional who examines pending transactions from a human perspective and makes a final decision.

[0486] "Final transaction processing" refers to the series of processes that approve or reject a transaction based on the reviewer's judgment.

[0487] "Notifications" are messages that inform users of the status or final outcome of a transaction.

[0488] The above definitions allow understanding of important terms related to this invention.

[0489] This invention is a system for preventing fraudulent remittances in digital payment services, specifically a system for real-time monitoring of transaction data and detection of abnormal patterns. The system includes major components such as users, terminals, and servers.

[0490] First, a user makes a remittance request using a device such as a smartphone or tablet. The user enters the recipient account information and the amount to be remitted, and transaction data is generated on the device. The generated transaction data includes the recipient information and the transaction amount.

[0491] The terminal encrypts the generated transaction data and then sends it to the server using a secure communication method (e.g., HTTPS). The server stores the transaction data received from the terminal in its system and verifies that the data is complete and accurate.

[0492] The server then inputs the received transaction data into a generative AI model, an algorithm that analyzes transaction data in real time against historical transaction data to detect anomalous transaction patterns. The AI ​​model identifies transactions that deviate from normal transaction patterns and marks transactions that are deemed potentially fraudulent as suspicious.

[0493] Any abnormal transaction detected will be temporarily suspended by the server and its status will be set to "Pending." At the same time, the server will send a notification to the user saying "The transaction is currently pending." This will let the user know that the transaction is pending and undergoing final confirmation.

[0494] The server then sends the pending transaction data to a human reviewer who examines the transaction details, checking the recipient account, time of day, amount, etc., to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[0495] If the reviewer approves the transaction, the server will re-promote the transaction, or if the transaction is rejected, it will cancel the transaction and send a "Transaction Cancelled" notification to the user.

[0496] For example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, where it is identified as an anomaly by the AI ​​model. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines there are no problems, the transaction is finally completed. If there are problems, the transaction is canceled and the user is notified.

[0497] An example of an input prompt for a generative AI model is:

[0498] Analyze the transaction data below to determine if there is any possibility of fraud.

[0499] Transaction Data:

[0500] User ID: 12345

[0501] Remittance destination: ABC Bank

[0502] Amount to be sent: 100,000 yen

[0503] Past trading patterns: Over 90% of transactions were under 10,000 yen

[0504] This transaction is unusual because it deviates significantly from the normal pattern.

[0505] This system effectively prevents fraudulent remittances in digital payment services, providing users with a service they can use with peace of mind. Also, through specific operational procedures, it is possible to understand in detail how this invention is implemented.

[0506] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0507] Step 1: Collect transaction data

[0508] The user makes a remittance request using the terminal. The user enters the remittance account information and the remittance amount, and transaction data is generated. This transaction data includes the remittance account information and the transaction amount.

[0509] (Input) The remittance destination information and transaction amount entered by the user into the terminal.

[0510] (Output) The generated transaction data.

[0511] Specific example of operation: The user opens the app, enters the recipient account information and the amount to send, and presses the "Send" button to generate transaction data.

[0512] Step 2: Sending transaction data

[0513] The terminal sends the generated transaction data to the server, which transmits the data using a secure communication method (e.g. HTTPS).

[0514] (Input) Transaction data generated by the terminal.

[0515] (Output) The encrypted transaction data sent to the server.

[0516] Specific operation example: The transaction data generated by the terminal is encrypted and sent to the server using the REST API.

[0517] Step 3: Receiving transaction data

[0518] The server receives the transaction data sent from the terminal, verifies that the data is complete and accurate, and stores the data in the system.

[0519] (Input) Encrypted transaction data sent from the terminal.

[0520] (Output) Transaction data stored in the system.

[0521] Specific operation example: The server saves the received data in a database and returns a reception confirmation response to the terminal.

[0522] Step 4: Real-time monitoring and analysis

[0523] The server inputs the received transaction data into a generative AI model, which analyzes the transaction data and detects anomalous patterns.

[0524] (Input) Transaction data stored on the server.

[0525] (Output) Trade valuation analyzed by the AI ​​model.

[0526] Specific example of operation: The server inputs transaction data into the generative AI model, and the AI ​​compares it with past transaction data and performs analysis.

[0527] Step 5: Detecting abnormal patterns

[0528] The generative AI model evaluates trading patterns against historical transaction data and marks transactions as suspicious if an abnormal trading pattern is detected.

[0529] (Input) Trade valuation analyzed by the AI ​​model.

[0530] (Output) Transactions marked as suspicious.

[0531] Specific example of operation: Transactions that the AI ​​determines to be abnormal are marked as "suspicious" and information about the suspicious transaction is recorded in a log.

[0532] Step 6: Suspicious Transaction Suspension

[0533] The server temporarily holds the suspicious transaction and sets its status to "Pending." The user is notified that "The transaction is currently pending."

[0534] (Input) Transactions marked as suspicious.

[0535] (Output) Transaction status set to "Pending" and user notification.

[0536] Example of specific behavior: The server updates the transaction status to "pending" and sends a pending notification to the user.

[0537] Step 7: Send to human reviewers

[0538] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[0539] (Input) "Pending" transaction data.

[0540] (Output) The transaction data sent to the reviewer.

[0541] Specific operation example: The server sends pending transaction data to the reviewer's system, and the reviewer receives the transaction data and starts reviewing it.

[0542] Step 8: Transaction verification and final confirmation

[0543] Human reviewers check the destination account, time of transaction, amount sent, etc. to determine whether the transaction is legitimate.

[0544] (Input) The transaction data received by the reviewer.

[0545] (Output) The transaction verification result.

[0546] Example of specific operation: A reviewer checks the remittance destination, transaction amount, etc., and asks the user for additional confirmation if necessary.

[0547] Step 9: Making a final decision

[0548] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0549] (Input) Transaction verification result.

[0550] (Output) Final processing results and user notification.

[0551] Example of specific operation: A reviewer makes a decision on the system to approve or reject a transaction, and the server proceeds with approved transactions, cancels rejected transactions, and sends a final notification to the user.

[0552] (Application example 1)

[0553] 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."

[0554] The lack of a system to effectively prevent fraudulent transfers is a problem for existing digital payment services. In particular, it is difficult to monitor transaction data in real time, detect anomalies, temporarily suspend suspicious transactions, and promptly notify users and reviewers. This can lead to a lack of swift action when fraudulent transactions occur, which can result in a loss of user trust.

[0555] 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.

[0556] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for analyzing the monitored transaction data and detecting abnormal transaction patterns, means for temporarily suspending the detected abnormal transactions, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, means for notifying the user of the final processing result, and means for automatically notifying the user and the reviewer by email when an abnormal transaction is detected. This makes it possible to detect abnormal transactions in real time and respond to them quickly and efficiently.

[0557] "Transaction data" refers to information relating to transfers and payments made by users, including details such as the transfer destination, transfer amount, time, and transaction ID.

[0558] "Real-time monitoring means" refers to methods and systems for capturing transaction data in real time and analyzing it without delay.

[0559] "Means for detecting abnormal trading patterns" refers to AI algorithms or rule-based systems that identify transactions that are unusual or risky compared to normal trading history.

[0560] "Means for temporarily suspending abnormal transactions" refers to a system that has the ability to immediately stop and suspend a suspicious transaction when it is detected.

[0561] "Means of submission to human reviewer" refers to the communications methods and protocols used to submit details of a pending transaction to a human reviewer for final review.

[0562] "Means for finalizing transaction processing based on reviewer's judgment" refers to the system functionality for executing or canceling a transaction based on the reviewer's evaluation and decision.

[0563] "Means for notifying the user of the final processing result" refers to the notification method used to inform the user of the result of a transaction after it has been completed or canceled.

[0564] "Means for automatic email notification" refers to a system function that automatically sends emails to users and reviewers based on pre-set templates when an anomaly in transaction data is detected.

[0565] The system for realizing the present invention is configured based on a program including the following processing procedures.

[0566] First, a user makes a remittance request using a terminal. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This transaction data is sent to the server in real time.

[0567] The server receives the received transaction data and begins monitoring it in real time. It uses AI algorithms such as IsolationForest to detect abnormal transaction patterns and identify transactions that are unusual or high risk compared to normal transaction history (e.g., when transactions of several thousand yen are usually made, but a large amount is suddenly transferred).

[0568] Transactions that are identified as anomalous are temporarily put on hold and the transaction status is set to "Pending." Details of the pending transaction are sent to the user and human reviewer using an automated email notification feature. The user is notified that the transaction is currently on hold, and the reviewer is provided with details of the transaction.

[0569] Next, a reviewer will examine the submitted transaction data to determine whether or not the transaction is abnormal. This review will take into account the user's past transaction history, remittance destination information, and the time of the transaction. Based on the reviewer's decision, the final transaction processing will be carried out. If the transaction is approved, it will proceed; if it is rejected, it will be canceled.

[0570] The final transaction processing result will be notified to the user. Notifications will include messages such as "Transaction completed" or "Transaction canceled" using an automated email notification function.

[0571] This system mainly uses the following hardware and software:

[0572] Hardware: Devices such as smartphones

[0573] Software: Python, pandas (data processing), scikit-learn (AI modeling), smtplib (email sending)

[0574] For example, if a user normally sends a few thousand yen, but suddenly attempts to send 150,000 yen, the AI ​​will detect this suspicious transaction and put it on hold. The user and reviewers will be notified, and the reviewers will determine the legitimacy of the transaction.

[0575] An example prompt is:

[0576] Please provide a detailed review of the following transaction.

[0577] User ID: 1

[0578] Amount to be sent: 150,000 yen

[0579] Timestamp: 2023-10-10 16:00:00

[0580] New Recipient: True

[0581] The transaction is currently pending. You have been notified and can now decide whether to accept or cancel the transaction.

[0582] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0583] Step 1:

[0584] A user makes a remittance request using a smartphone. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This generates transaction data. The inputs are the "remittance amount," "remittance destination information," and "timestamp," and the output is "transaction data."

[0585] Step 2:

[0586] The terminal sends the generated transaction data to the server. Based on the transaction data, the server receives the data. The input here is "transaction data" and the output is "transaction data received by server."

[0587] Step 3:

[0588] The server begins monitoring transaction data in real time. The received transaction data is input into an AI algorithm such as IsolationForest, which analyzes it to detect abnormal transaction patterns. The input is "transaction data" and the output is "detection results of abnormal patterns." Specifically, each field of the transaction data (such as the amount sent and the recency of the recipient) is applied to the AI ​​algorithm.

[0589] Step 4:

[0590] Transactions that are determined to be abnormal are temporarily suspended and the transaction status is set to "pending." This suspension process causes the server to mark the abnormal transaction as a flag and handle it specially in subsequent processes. The input is the "detection result of an abnormal transaction," and the output is the "pending transaction data."

[0591] Step 5:

[0592] The server automatically sends emails with details of the released transaction to the user and reviewer. Specifically, emails are sent using smtplib. The input is "pending transaction data" and the output is "sent email."

[0593] Step 6:

[0594] The reviewer scrutinizes the transaction data received by email and determines whether or not the transaction is anomalous. The input here is the "transaction data received by email" and the output is the "review results." Specifically, the reviewer determines the validity of the anomalous transaction based on the transaction history, remittance destination information, transaction time period, etc.

[0595] Step 7:

[0596] Based on the reviewer's decision, the server performs the final transaction processing. If the transaction is approved, the transaction data is processed as usual. If it is rejected, the transaction is canceled. The input is the "review result" and the output is the "final transaction processing result."

[0597] Step 8:

[0598] The server notifies the user of the final processing result. This notification is sent via automatic email using smtplib. The input is the "final transaction processing result" and the output is "notification to the user." Specifically, an email is sent to the user stating the status of the transaction, whether it was approved or canceled.

[0599] 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.

[0600] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[0601] A natural language description of what the program does

[0602] 1. Collection of transaction data

[0603] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[0604] 2. Transmission of transaction data

[0605] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[0606] 3. Receiving Transaction Data

[0607] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it also starts the user's emotion engine.

[0608] 4. Acquiring Emotion Data

[0609] It captures emotional data about users through cameras, microphones, and other sensors built into their devices, such as facial expressions, tone of voice, and stress levels.

[0610] 5. Emotion Data Analysis

[0611] The emotion engine analyzes the acquired emotion data and determines the user's emotional state (for example, impatience, tension, relief).

[0612] 6. Real-time monitoring and analysis

[0613] The server then passes the received transaction data to the AI ​​system for real-time monitoring and analysis, which compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns.

[0614] 7. Detecting abnormal patterns and emotional states

[0615] The AI ​​system will mark a transaction as suspicious if it detects an abnormal trading pattern, and an emotional engine will make an additional determination as to whether a transaction is anomalous based on the user's emotional state.

[0616] 8. Suspicious Transaction Suspension

[0617] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[0618] 9. Send to human reviewers

[0619] The server sends the pending transaction data and associated sentiment data to human reviewers, who then scrutinize the data.

[0620] 10. Transaction Verification and Final Confirmation

[0621] Human reviewers consider the destination account, transaction time, transaction amount, and captured emotional state to make the final decision.

[0622] 11. Implementation of Final Decision

[0623] If the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0624] To give a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they are deemed legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[0625] This system will effectively prevent fraudulent transfers and irregular transactions in digital payment services, providing users with a service they can use with greater peace of mind.

[0626] The processing flow will be explained below.

[0627] Step 1:

[0628] The user inputs a remittance request using the terminal, specifies the recipient's account information and the amount to be remitted, and presses the remit button.

[0629] Step 2:

[0630] The terminal sends the generated transaction data (recipient account information, remittance amount, user ID, transaction time) to the server.

[0631] Step 3:

[0632] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it starts the emotion engine and starts acquiring the user's emotion data.

[0633] Step 4:

[0634] The device collects emotional data from users through built-in cameras, microphones, and other sensors, such as facial expressions, tone of voice, and heart rate.

[0635] Step 5:

[0636] The device sends the collected emotion data to the server, which then passes the data to the emotion engine.

[0637] Step 6:

[0638] The emotion engine analyzes the transmitted emotion data and determines the user's emotional state (e.g., anxious, nervous, relieved).

[0639] Step 7:

[0640] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with past transaction history and statistical data to detect abnormal patterns.

[0641] Step 8:

[0642] If the AI ​​system detects an abnormal transaction pattern, for example, if an amount that is excessively high compared to the amount normally sent, it will notify the server.

[0643] Step 9:

[0644] The server combines the results of the AI ​​system and the analysis results of the emotion engine to make a final decision on whether a transaction is suspicious. For example, if a transaction is abnormal and the user is in a tense state, the server will determine that the transaction is suspicious.

[0645] Step 10:

[0646] The server will temporarily suspend the suspicious transaction, set the transaction status to "Pending," and send a notification to the user stating, "The transaction is currently pending."

[0647] Step 11:

[0648] The server sends the pending transaction data and associated emotion data to a human reviewer, who is provided with the transaction details and emotional state.

[0649] Step 12:

[0650] Human reviewers will make a final decision based on the destination account, transaction time, transaction amount, and captured emotional data, taking into account, for example, if the transaction amount is significantly larger than normal or if the user appears to be in an unnatural state of stress.

[0651] Step 13:

[0652] If the reviewer approves the transaction, they send the approval result to the server, which updates the transaction status to "approved" and allows the transaction to proceed.

[0653] Step 14:

[0654] If the transaction is approved, the server proceeds with the actual transfer process. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[0655] Step 15:

[0656] If the reviewer decides to reject the transaction, the rejection is sent to the server, which updates the transaction status to "Cancelled" and cancels the transaction.

[0657] Step 16:

[0658] If the transaction is cancelled, the server sends a notification to the user's terminal that the transaction has been cancelled.

[0659] In this way, this system can combine and analyze transaction data and emotional data to detect and prevent suspicious transactions with high accuracy.

[0660] Example 2

[0661] 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."

[0662] In recent years, fraudulent remittances and abnormal transactions in digital payment services have been increasing, and conventional security measures are often inadequate to address these issues. New technologies are needed to detect fraudulent activity in advance and respond quickly. Furthermore, taking the user's emotional state into account is expected to enable more accurate detection of abnormal transactions, but achieving this requires advanced analytical technology and real-time processing. The present invention aims to solve these problems.

[0663] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving transaction data, means for monitoring the transaction data in real time, means for analyzing the transaction data and detecting abnormal transaction patterns, means for acquiring and analyzing user emotion data to identify the user's emotional state, means for temporarily suspending abnormal transactions based on the abnormal transactions and the emotion analysis results, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing results. This makes it possible to detect fraudulent transactions with high accuracy from both the transaction data and the emotion data, and to prevent fraud quickly and reliably.

[0664] "Transaction data" refers to data that includes detailed information such as the recipient account information specified by the user, the amount to be transferred, the user ID, and the time of the transaction.

[0665] "Real-time monitoring" means that as soon as transaction data is generated, its contents are immediately investigated and processing is carried out to detect any fraud or abnormalities.

[0666] "Abnormal trading patterns" refer to trading patterns that are clearly different from normal trading behavior or past trading history, or that are suspected of being fraudulent.

[0667] "Emotional data" refers to data regarding a user's emotional state, such as a user's facial expression, tone of voice, stress level, etc.

[0668] An "emotion engine" refers to software or algorithms that analyze emotional data and identify a user's emotional state.

[0669] "Temporary suspension" means that the abnormal transaction detected will not be immediately processed, but will be temporarily stopped for further inspection and confirmation.

[0670] "Reviewer" refers to a human professional who reviews transaction data and sentiment data to make a final determination regarding suspicious transactions.

[0671] "Final transaction processing" refers to the process of proceeding with or canceling a transfer based on the reviewer's determination of whether the transaction is fraudulent or not.

[0672] "Notification" refers to messages or information that inform a user about the status of a transaction or the final processing result.

[0673] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[0674] First, the user opens the application on the terminal and initiates a transaction. They enter a transfer request and specify the recipient's account information and the amount to be transferred. When the user presses the "Transfer" button, this information is generated as transaction data by the terminal. The terminal then encrypts the generated transaction data and sends it to the server via a secure network. The transaction data includes details such as the recipient's account information, the amount to be transferred, the user ID, and the time of the transaction.

[0675] The server receives transaction data sent from the terminal in real time and stores it in a database. The server also sends commands to the user's device to activate the camera, microphone, and other sensors to obtain the user's emotional data. The device collects the user's facial expressions, tone of voice, stress level, and other information in real time and generates emotional data. This emotional data is sent back to the server and analyzed by an emotion engine to identify the user's emotional state. For example, it determines states such as "anxiety" or "tension."

[0676] The server then passes the received transaction data to an AI system for real-time monitoring and analysis. The AI ​​system compares the transaction data with past transaction patterns and statistical data to detect abnormal transaction patterns. Based on the detected abnormal transactions and the results of sentiment analysis, the system temporarily suspends suspicious transactions and sets their status to "on hold." The server then sends a notification to the user stating, "The transaction is currently on hold."

[0677] The server then sends the pending transaction data and associated emotion data to a human reviewer who reviews the data and makes a final decision. The human reviewer reviews the destination account, transaction time, transaction amount, and captured emotion state. If necessary, the human reviewer can provide additional confirmation to the user.

[0678] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0679] This system effectively prevents fraudulent transfers and irregular transactions in digital payment services, allowing users to use the service with peace of mind.

[0680] As a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​system determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they determine that it is legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[0681] Prompt Sentence Examples

[0682] "Describe a scenario where a user is attempting to transfer a large amount of money. Detail how you would process it by analyzing sentiment data and detecting anomalous transactions."

[0683] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0684] Step 1:

[0685] The user opens the application on the terminal and enters a remittance request. The data entered includes the recipient's account information, the remittance amount, the user ID, and the transaction time. When the user presses the "Remit" button, this data is generated as transaction data by the terminal. The terminal then encrypts this transaction data and sends it to the server over a secure network.

[0686] Input: Remittance account information, remittance amount, user ID, transaction time

[0687] Output: Encrypted transaction data

[0688] Step 2:

[0689] The server receives the transaction data sent from the terminal and stores it in a database. The server also starts the emotion engine and prepares the user's device to collect emotion data.

[0690] Input: Encrypted transaction data

[0691] Output: Transaction data stored in a database

[0692] Step 3:

[0693] The server sends commands to the user's device to activate the camera, microphone, and other sensors, and the device collects the user's facial expressions, tone of voice, stress level, and other information in real time to generate emotional data.

[0694] Input: Commands to the user device

[0695] Output: User emotion data

[0696] Step 4:

[0697] Emotional data is sent from the user's device to the server, which then analyzes it with an emotion engine to identify the user's emotional state. For example, it determines whether the user is feeling anxious or nervous.

[0698] Input: Emotion data

[0699] Output: Parsed emotional state

[0700] Step 5:

[0701] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with historical transaction patterns and statistical data to detect abnormal transaction patterns.

[0702] Input: Transaction data

[0703] Output: Abnormal trading pattern detection results

[0704] Step 6:

[0705] If the AI ​​system detects an abnormal trading pattern, it will mark the transaction as suspicious, and if the emotion engine determines that the transaction is suspicious based on the user's emotional state, the transaction will also be marked as suspicious.

[0706] Input: Detected abnormal trading patterns, analyzed emotional state

[0707] Output: Suspicious transaction mark

[0708] Step 7:

[0709] The server temporarily holds the suspicious transaction and sets its status to "Pending." The server sends a notification to the user saying, "The transaction is currently pending."

[0710] Input: Mark suspicious transaction

[0711] Output: Pending transaction status, notification to user

[0712] Step 8:

[0713] The server sends the pending transaction data and associated sentiment data to human reviewers, who scrutinize the data and check for anomalies.

[0714] Input: Pending transaction data, associated sentiment data

[0715] Output: Send data to reviewers

[0716] Step 9:

[0717] A human reviewer will review the details of the destination account, transaction time, transaction amount, and captured emotional state. If necessary, the reviewer will ask the user for additional confirmation.

[0718] Input: Data to reviewer

[0719] Output: Final decision

[0720] Step 10:

[0721] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0722] Input: Final decision

[0723] Output: Transaction proceeds or is cancelled, and the user is notified

[0724] (Application example 2)

[0725] 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."

[0726] Increasing fraudulent remittances and abnormal transactions are becoming a major problem in modern digital payment services. This puts users at risk of fraudulent transactions and may undermine trust in payment services. Furthermore, conventional systems detect anomalies solely by analyzing transaction data, which means there is a risk of overlooking fraudulent transactions. Furthermore, accurate judgments are difficult because they do not take into account the user's emotional state. Therefore, a more accurate fraud detection system is needed.

[0727] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0728] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for acquiring user emotional data, means for analyzing the acquired emotional data to determine the user's emotional state, means for analyzing the monitored transaction data and temporarily suspending transactions determined to be anomalous based on anomalous transaction patterns and the user's emotional state, means for transmitting the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This makes it possible to improve the accuracy of detecting anomalous transactions and effectively reduce the risk of fraudulent remittances by taking the user's emotional state into consideration in addition to analyzing the transaction data.

[0729] "Transaction data" is information generated when a user transfers or makes a payment, and includes details such as the recipient's account information, the amount transferred, the user ID, and the time of the transaction.

[0730] An "abnormal transaction pattern" is a transaction that shows unnatural trends that differ from normal transactions and is an indicator of possible fraudulent transfers or other fraudulent activity.

[0731] "Emotion data" is data that reflects the user's emotional state, obtained by analyzing the user's facial expressions, tone of voice, stress level, etc.

[0732] The "emotional state" indicates the psychological state of the user, and refers to states such as impatience, tension, and relief.

[0733] "Temporary suspension" refers to the process of temporarily halting a transaction that has been determined to be abnormal and withholding a final decision.

[0734] A "human reviewer" is someone who manually reviews transactions and sentiment data held by the system and ultimately approves or rejects the transaction.

[0735] The system that realizes this application example consists of the following steps: First, a user launches a digital payment service application using a smartphone terminal. When the user enters a remittance request and specifies the recipient's account information and the remittance amount, the smartphone terminal generates transaction data and sends it to the server.

[0736] The server stores the received transaction data in a database. The transaction data includes details such as the recipient's account information, the remittance amount, the user ID, and the transaction time. At the same time, the user's emotion engine runs on the smartphone and collects the user's emotion data through the device's camera and microphone.

[0737] Emotion data includes the user's facial expressions, tone of voice, stress level, etc. The emotion engine analyzes the emotion data in real time to determine the user's emotional state. This emotion data is then sent to the server and integrated with the transaction data.

[0738] After integrating the above data, the server's AI system uses the received transaction data and emotional state data for real-time monitoring and analysis. The AI ​​system compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns, and then takes the emotional state into account to make a final anomalous transaction judgment. Transactions that are judged to be abnormal are temporarily suspended and sent to human reviewers for a final decision.

[0739] The reviewer verifies the validity of the transaction based on the pending transaction data and the user's emotional state data sent from the server. Based on the review results, the server performs the final transaction processing and notifies the user of the result.

[0740] The specific hardware and software used are as follows:

[0741] Smartphone device: camera, microphone, processing unit

[0742] Server: Database management system (MySQL, etc.), AI model (TensorFlow / PyTorch)

[0743] Emotion analysis engine: Voice analysis library (Librosa, etc.), face recognition library (OpenCV, etc.)

[0744] Monitoring system: System monitoring tools (Prometheus, etc.)

[0745] As a concrete example, consider a situation where a user attempts to transfer more than 100,000 yen. If the AI ​​detects an abnormality during the transfer process, such as the camera detecting a panicked expression or the microphone detecting a tense tone of voice, the transaction will be temporarily put on hold. The server then sends this data to a human reviewer, who will then approve or reject the transaction based on their final judgment.

[0746] Example prompt sentence:

[0747] "If a user attempts to send more than 100,000 yen using a smartphone app, consider a system that verifies the intention of the transfer. AI will monitor abnormal transactions and the user's facial expressions in real time to determine safety."

[0748] This invention improves the accuracy of detecting abnormal transactions by combining analysis of transaction data with data on the user's emotional state, making it possible to effectively reduce the risk of fraudulent remittances.

[0749] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0750] Step 1:

[0751] A user launches a digital payment service application on a smartphone terminal. The user enters a remittance request, specifying the recipient's account information and the amount to be sent. When the "Remit" button is pressed, the terminal generates transaction data.

[0752] Input: Recipient account information, remittance amount, user ID, transaction time

[0753] Output: Transaction data

[0754] Step 2:

[0755] The terminal transmits the generated transaction data to the server.

[0756] Input: Transaction data

[0757] Output: Transaction data sent to the server

[0758] Step 3:

[0759] The server stores the received transaction data in a database.

[0760] Input: Transaction data sent from the terminal

[0761] Output: Transaction data stored in a database

[0762] Step 4:

[0763] At the same time, the device's camera and microphone are used to collect the user's emotional data: the camera captures the user's facial expressions, and the microphone captures the tone of voice and stress level.

[0764] Input: Camera video, audio data

[0765] Output: Emotion data

[0766] Step 5:

[0767] The device analyzes the collected emotional data and determines the user's emotional state. Analysis is performed using a face recognition library (such as OpenCV) and a voice analysis library (such as Librosa).

[0768] Input: Emotion data

[0769] Output: Emotional state (e.g., anxious, nervous, relieved)

[0770] Step 6:

[0771] The device transmits the analyzed emotional state to the server.

[0772] Input: Parsed emotional state

[0773] Output: Emotional state sent to the server

[0774] Step 7:

[0775] The server combines the received transaction data with the emotional state data and passes it to the AI ​​system.

[0776] Input: Transaction data, emotional state data

[0777] Output: Integrated data (transaction data + emotional state data)

[0778] Step 8:

[0779] The AI ​​system monitors and analyzes the integrated data in real time, comparing it with previous trading patterns and statistical data to detect abnormal patterns, and then takes into account emotional states to make a final judgment on abnormal trading.

[0780] Inputs: Consolidated data, historical trading patterns, statistical data

[0781] Output: Abnormal transaction determination result

[0782] Step 9:

[0783] The server temporarily holds transactions that are identified as abnormal and sends the transaction data and emotional state data to a human reviewer.

[0784] Input: Abnormal transaction judgment results, transaction data, emotional state data

[0785] Output: Pending transactions, reviewer notifications

[0786] Step 10:

[0787] The reviewer evaluates the validity of the transaction based on the pending transaction data and emotional state data. If the transaction is deemed valid, the server proceeds with the transaction, but if there are any problems, the server cancels the transaction.

[0788] Input: Pending transaction data, emotional state data

[0789] Output: Review result (approved or rejected)

[0790] Step 11:

[0791] The server performs the final transaction processing based on the review results and notifies the user of the results.

[0792] Input: Review results

[0793] Output: Notification of final transaction result (transfer completion or transaction cancellation notification)

[0794] 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.

[0795] 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.

[0796] 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.

[0797] [Third embodiment]

[0798] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0799] 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.

[0800] 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).

[0801] 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.

[0802] 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.

[0803] 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).

[0804] 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.

[0805] 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.

[0806] 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.

[0807] 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.

[0808] 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.

[0809] 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."

[0810] This invention is a system for preventing fraudulent remittances in digital payment services. It uses AI to monitor transaction data in real time and detect abnormal transaction patterns. Furthermore, if a suspicious transaction is detected, the system temporarily suspends the transaction and leaves the final decision to a human reviewer, making a more accurate judgment.

[0811] A natural language description of what the program does

[0812] 1. Collection of transaction data

[0813] A user makes a remittance request from a terminal. For example, if a user attempts to remit 100,000 yen, transaction data (remittance destination information and transaction amount) is generated.

[0814] 2. Transmission of transaction data

[0815] The terminal transmits the generated transaction data to the server.

[0816] 3. Receiving Transaction Data

[0817] The server receives the transaction data sent from the terminal.

[0818] 4. Real-time monitoring and analysis

[0819] The server inputs the received transaction data into an AI algorithm, which analyzes the transaction data and detects abnormal patterns. For example, if a user who usually makes transactions of less than 10,000 yen suddenly transfers 100,000 yen, this would be an abnormal pattern.

[0820] 5. Detecting Abnormal Patterns

[0821] The AI ​​algorithm evaluates trading patterns against a user's past trading history and marks the transaction as suspicious if an abnormal trading pattern is detected.

[0822] 6. Suspicious Transaction Suspension

[0823] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[0824] 7. Send to human reviewers

[0825] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[0826] 8. Transaction Verification and Final Confirmation

[0827] Human reviewers check the recipient account, time of transaction, amount of transfer, etc. to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[0828] 9. Implementation of the Final Decision

[0829] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0830] To give a specific example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server and the AI ​​determines that it is an anomaly. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines that there are no problems, the transaction is finally completed. If there is a problem, the transaction is canceled and the user is notified.

[0831] This system effectively prevents fraudulent transfers via digital payment services, providing users with a service they can use with peace of mind.

[0832] The processing flow will be explained below.

[0833] Step 1:

[0834] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[0835] Step 2:

[0836] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[0837] Step 3:

[0838] The server receives the transaction data sent from the terminal and stores it in a database.

[0839] Step 4:

[0840] The server passes the transaction data to an AI system for real-time monitoring and analysis, which uses algorithms to analyze the transaction data and detect abnormal transaction patterns.

[0841] Step 5:

[0842] The AI ​​system compares transaction data with past transaction history and statistical data to assess whether there are any unusual patterns, for example detecting unusually high transfer amounts compared to the average of a user's past transactions.

[0843] Step 6:

[0844] If the AI ​​system detects an abnormal transaction pattern, the server marks the transaction as suspicious and puts it on hold, at which point the transaction's status is set to "pending."

[0845] Step 7:

[0846] The server sends a notification to the user's terminal that "The transaction is currently pending."

[0847] Step 8:

[0848] The server sends the pending transaction data to a human reviewer, who is provided with transaction details (destination account, transaction amount, user ID, etc.).

[0849] Step 9:

[0850] Human reviewers scrutinize transaction details, such as checking the trustworthiness of the recipient and whether similar transactions have been made before.

[0851] Step 10:

[0852] If the reviewer approves the transaction, the approval result is sent to the server, and the transaction status is updated to "Approved."

[0853] Step 11:

[0854] The server resumes the approved transaction and proceeds with the actual transfer. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[0855] Step 12:

[0856] If the reviewer decides to reject the transaction, they send the rejection result to the server, and the transaction status is set to "Cancelled" because the transaction was rejected.

[0857] Step 13:

[0858] The server cancels the transaction and sends a "transaction canceled" notification to the user's terminal.

[0859] This allows the system to detect abnormal transactions in real time and take appropriate action to prevent fraudulent transfers.

[0860] Example 1

[0861] 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."

[0862] Conventional digital payment systems have difficulty detecting sophisticated fraudulent remittances, preventing users from conducting transactions with peace of mind. Furthermore, they lacked mechanisms for detecting abnormal transaction patterns in real time and taking appropriate countermeasures. This increased the risk of fraudulent transactions, potentially resulting in significant losses for financial institutions and users. This invention aims to solve these issues by utilizing AI for advanced monitoring and analysis, effectively preventing fraudulent transactions.

[0863] 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.

[0864] In this invention, the server includes means for receiving transaction data sent by a user from a terminal, means for transmitting the received transaction data to the server via a secure communication means, means for storing the transaction data received by the server within the system, means for inputting the received transaction data into a generative AI model, means for monitoring and analyzing the transaction data in real time using the generative AI model, means for detecting anomalous transaction patterns from the analyzed transaction data, means for temporarily suspending the detected anomalous transaction and setting its status to "suspended," means for notifying the user of the suspended transaction data, means for transmitting the suspended transaction to a human reviewer, means for the reviewer to carefully examine the transaction content and make a judgment, means for final transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This enables real-time monitoring of transaction data and early detection of anomalies, thereby significantly reducing the risk of fraudulent transactions.

[0865] "Transaction data" is a collection of information related to remittances and payments made by users, including details such as the recipient account information and the remittance amount.

[0866] "Terminal" refers to the digital device used by a user, such as a smartphone or tablet, that provides the interface for making a remittance request.

[0867] A "server" is a computer system that receives, processes, and analyzes transaction data sent from a terminal.

[0868] A "generative AI model" is a model that uses artificial intelligence to analyze trading data and includes algorithms to detect anomalous trading patterns.

[0869] "Real-time monitoring" is the process of monitoring transaction data in real time and analyzing it immediately.

[0870] "Abnormal transaction patterns" refer to potentially fraudulent transaction behavior that deviates from normal transaction patterns and is detected by the AI ​​model.

[0871] "Suspicious transactions" are transactions that are marked as anomalous by the AI ​​model and require further review.

[0872] "Suspension" refers to a state in which a transaction in which an abnormality has been detected is temporarily halted and a final decision is pending.

[0873] A "reviewer" is a professional who examines pending transactions from a human perspective and makes a final decision.

[0874] "Final transaction processing" refers to the series of processes that approve or reject a transaction based on the reviewer's judgment.

[0875] "Notifications" are messages that inform users of the status or final outcome of a transaction.

[0876] The above definitions allow understanding of important terms related to this invention.

[0877] This invention is a system for preventing fraudulent remittances in digital payment services, specifically a system for real-time monitoring of transaction data and detection of abnormal patterns. The system includes major components such as users, terminals, and servers.

[0878] First, a user makes a remittance request using a device such as a smartphone or tablet. The user enters the recipient account information and the amount to be remitted, and transaction data is generated on the device. The generated transaction data includes the recipient information and the transaction amount.

[0879] The terminal encrypts the generated transaction data and then sends it to the server using a secure communication method (e.g., HTTPS). The server stores the transaction data received from the terminal in its system and verifies that the data is complete and accurate.

[0880] The server then inputs the received transaction data into a generative AI model, an algorithm that analyzes transaction data in real time against historical transaction data to detect anomalous transaction patterns. The AI ​​model identifies transactions that deviate from normal transaction patterns and marks transactions that are deemed potentially fraudulent as suspicious.

[0881] Any abnormal transaction detected will be temporarily suspended by the server and its status will be set to "Pending." At the same time, the server will send a notification to the user saying "The transaction is currently pending." This will let the user know that the transaction is pending and undergoing final confirmation.

[0882] The server then sends the pending transaction data to a human reviewer who examines the transaction details, checking the recipient account, time of day, amount, etc., to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[0883] If the reviewer approves the transaction, the server will re-promote the transaction, or if the transaction is rejected, it will cancel the transaction and send a "Transaction Cancelled" notification to the user.

[0884] For example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, where it is identified as an anomaly by the AI ​​model. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines there are no problems, the transaction is finally completed. If there are problems, the transaction is canceled and the user is notified.

[0885] An example of an input prompt for a generative AI model is:

[0886] Analyze the transaction data below to determine if there is any possibility of fraud.

[0887] Transaction Data:

[0888] User ID: 12345

[0889] Remittance destination: ABC Bank

[0890] Amount to be sent: 100,000 yen

[0891] Past trading patterns: Over 90% of transactions were under 10,000 yen

[0892] This transaction is unusual because it deviates significantly from the normal pattern.

[0893] This system effectively prevents fraudulent remittances in digital payment services, providing users with a service they can use with peace of mind. Also, through specific operational procedures, it is possible to understand in detail how this invention is implemented.

[0894] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0895] Step 1: Collect transaction data

[0896] The user makes a remittance request using the terminal. The user enters the remittance account information and the remittance amount, and transaction data is generated. This transaction data includes the remittance account information and the transaction amount.

[0897] (Input) The remittance destination information and transaction amount entered by the user into the terminal.

[0898] (Output) The generated transaction data.

[0899] Specific example of operation: The user opens the app, enters the recipient account information and the amount to send, and presses the "Send" button to generate transaction data.

[0900] Step 2: Sending transaction data

[0901] The terminal sends the generated transaction data to the server, which transmits the data using a secure communication method (e.g. HTTPS).

[0902] (Input) Transaction data generated by the terminal.

[0903] (Output) The encrypted transaction data sent to the server.

[0904] Specific operation example: The transaction data generated by the terminal is encrypted and sent to the server using the REST API.

[0905] Step 3: Receiving transaction data

[0906] The server receives the transaction data sent from the terminal, verifies that the data is complete and accurate, and stores the data in the system.

[0907] (Input) Encrypted transaction data sent from the terminal.

[0908] (Output) Transaction data stored in the system.

[0909] Specific operation example: The server saves the received data in a database and returns a reception confirmation response to the terminal.

[0910] Step 4: Real-time monitoring and analysis

[0911] The server inputs the received transaction data into a generative AI model, which analyzes the transaction data and detects anomalous patterns.

[0912] (Input) Transaction data stored on the server.

[0913] (Output) Trade valuation analyzed by the AI ​​model.

[0914] Specific example of operation: The server inputs transaction data into the generative AI model, and the AI ​​compares it with past transaction data and performs analysis.

[0915] Step 5: Detecting abnormal patterns

[0916] The generative AI model evaluates trading patterns against historical transaction data and marks transactions as suspicious if an abnormal trading pattern is detected.

[0917] (Input) Trade valuation analyzed by the AI ​​model.

[0918] (Output) Transactions marked as suspicious.

[0919] Specific example of operation: Transactions that the AI ​​determines to be abnormal are marked as "suspicious" and information about the suspicious transaction is recorded in a log.

[0920] Step 6: Suspicious Transaction Suspension

[0921] The server temporarily holds the suspicious transaction and sets its status to "Pending." The user is notified that "The transaction is currently pending."

[0922] (Input) Transactions marked as suspicious.

[0923] (Output) Transaction status set to "Pending" and user notification.

[0924] Example of specific behavior: The server updates the transaction status to "pending" and sends a pending notification to the user.

[0925] Step 7: Send to human reviewers

[0926] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[0927] (Input) "Pending" transaction data.

[0928] (Output) The transaction data sent to the reviewer.

[0929] Specific operation example: The server sends pending transaction data to the reviewer's system, and the reviewer receives the transaction data and starts reviewing it.

[0930] Step 8: Transaction verification and final confirmation

[0931] Human reviewers check the destination account, time of transaction, amount sent, etc. to determine whether the transaction is legitimate.

[0932] (Input) The transaction data received by the reviewer.

[0933] (Output) The transaction verification result.

[0934] Example of specific operation: A reviewer checks the remittance destination, transaction amount, etc., and asks the user for additional confirmation if necessary.

[0935] Step 9: Making a final decision

[0936] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[0937] (Input) Transaction verification result.

[0938] (Output) Final processing results and user notification.

[0939] Example of specific operation: A reviewer makes a decision on the system to approve or reject a transaction, and the server proceeds with approved transactions, cancels rejected transactions, and sends a final notification to the user.

[0940] (Application example 1)

[0941] 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."

[0942] The lack of a system to effectively prevent fraudulent transfers is a problem for existing digital payment services. In particular, it is difficult to monitor transaction data in real time, detect anomalies, temporarily suspend suspicious transactions, and promptly notify users and reviewers. This can lead to a lack of swift action when fraudulent transactions occur, which can result in a loss of user trust.

[0943] 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.

[0944] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for analyzing the monitored transaction data and detecting abnormal transaction patterns, means for temporarily suspending the detected abnormal transactions, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, means for notifying the user of the final processing result, and means for automatically notifying the user and the reviewer by email when an abnormal transaction is detected. This makes it possible to detect abnormal transactions in real time and respond to them quickly and efficiently.

[0945] "Transaction data" refers to information relating to transfers and payments made by users, including details such as the transfer destination, transfer amount, time, and transaction ID.

[0946] "Real-time monitoring means" refers to methods and systems for capturing transaction data in real time and analyzing it without delay.

[0947] "Means for detecting abnormal trading patterns" refers to AI algorithms or rule-based systems that identify transactions that are unusual or risky compared to normal trading history.

[0948] "Means for temporarily suspending abnormal transactions" refers to a system that has the ability to immediately stop and suspend a suspicious transaction when it is detected.

[0949] "Means of submission to human reviewer" refers to the communications methods and protocols used to submit details of a pending transaction to a human reviewer for final review.

[0950] "Means for finalizing transaction processing based on reviewer's judgment" refers to the system functionality for executing or canceling a transaction based on the reviewer's evaluation and decision.

[0951] "Means for notifying the user of the final processing result" refers to the notification method used to inform the user of the result of a transaction after it has been completed or canceled.

[0952] "Means for automatic email notification" refers to a system function that automatically sends emails to users and reviewers based on pre-set templates when an anomaly in transaction data is detected.

[0953] The system for realizing the present invention is configured based on a program including the following processing procedures.

[0954] First, a user makes a remittance request using a terminal. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This transaction data is sent to the server in real time.

[0955] The server receives the received transaction data and begins monitoring it in real time. It uses AI algorithms such as IsolationForest to detect abnormal transaction patterns and identify transactions that are unusual or high risk compared to normal transaction history (e.g., when transactions of several thousand yen are usually made, but a large amount is suddenly transferred).

[0956] Transactions that are identified as anomalous are temporarily put on hold and the transaction status is set to "Pending." Details of the pending transaction are sent to the user and human reviewer using an automated email notification feature. The user is notified that the transaction is currently on hold, and the reviewer is provided with details of the transaction.

[0957] Next, a reviewer will examine the submitted transaction data to determine whether or not the transaction is abnormal. This review will take into account the user's past transaction history, remittance destination information, and the time of the transaction. Based on the reviewer's decision, the final transaction processing will be carried out. If the transaction is approved, it will proceed; if it is rejected, it will be canceled.

[0958] The final transaction processing result will be notified to the user. Notifications will include messages such as "Transaction completed" or "Transaction canceled" using an automated email notification function.

[0959] This system mainly uses the following hardware and software:

[0960] Hardware: Devices such as smartphones

[0961] Software: Python, pandas (data processing), scikit-learn (AI modeling), smtplib (email sending)

[0962] For example, if a user normally sends a few thousand yen, but suddenly attempts to send 150,000 yen, the AI ​​will detect this suspicious transaction and put it on hold. The user and reviewers will be notified, and the reviewers will determine the legitimacy of the transaction.

[0963] An example prompt is:

[0964] Please provide a detailed review of the following transaction.

[0965] User ID: 1

[0966] Amount to be sent: 150,000 yen

[0967] Timestamp: 2023-10-10 16:00:00

[0968] New Recipient: True

[0969] The transaction is currently pending. You have been notified and can now decide whether to accept or cancel the transaction.

[0970] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0971] Step 1:

[0972] A user makes a remittance request using a smartphone. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This generates transaction data. The inputs are the "remittance amount," "remittance destination information," and "timestamp," and the output is "transaction data."

[0973] Step 2:

[0974] The terminal sends the generated transaction data to the server. Based on the transaction data, the server receives the data. The input here is "transaction data" and the output is "transaction data received by server."

[0975] Step 3:

[0976] The server begins monitoring transaction data in real time. The received transaction data is input into an AI algorithm such as IsolationForest, which analyzes it to detect abnormal transaction patterns. The input is "transaction data" and the output is "detection results of abnormal patterns." Specifically, each field of the transaction data (such as the amount sent and the recency of the recipient) is applied to the AI ​​algorithm.

[0977] Step 4:

[0978] Transactions that are determined to be abnormal are temporarily suspended and the transaction status is set to "pending." This suspension process causes the server to mark the abnormal transaction as a flag and handle it specially in subsequent processes. The input is the "detection result of an abnormal transaction," and the output is the "pending transaction data."

[0979] Step 5:

[0980] The server automatically sends emails with details of the released transaction to the user and reviewer. Specifically, emails are sent using smtplib. The input is "pending transaction data" and the output is "sent email."

[0981] Step 6:

[0982] The reviewer scrutinizes the transaction data received by email and determines whether or not the transaction is anomalous. The input here is the "transaction data received by email" and the output is the "review results." Specifically, the reviewer determines the validity of the anomalous transaction based on the transaction history, remittance destination information, transaction time period, etc.

[0983] Step 7:

[0984] Based on the reviewer's decision, the server performs the final transaction processing. If the transaction is approved, the transaction data is processed as usual. If it is rejected, the transaction is canceled. The input is the "review result" and the output is the "final transaction processing result."

[0985] Step 8:

[0986] The server notifies the user of the final processing result. This notification is sent via automatic email using smtplib. The input is the "final transaction processing result" and the output is "notification to the user." Specifically, an email is sent to the user stating the status of the transaction, whether it was approved or canceled.

[0987] 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.

[0988] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[0989] A natural language description of what the program does

[0990] 1. Collection of transaction data

[0991] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[0992] 2. Transmission of transaction data

[0993] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[0994] 3. Receiving Transaction Data

[0995] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it also starts the user's emotion engine.

[0996] 4. Acquiring Emotion Data

[0997] It captures emotional data about users through cameras, microphones, and other sensors built into their devices, such as facial expressions, tone of voice, and stress levels.

[0998] 5. Emotion Data Analysis

[0999] The emotion engine analyzes the acquired emotion data and determines the user's emotional state (for example, impatience, tension, relief).

[1000] 6. Real-time monitoring and analysis

[1001] The server then passes the received transaction data to the AI ​​system for real-time monitoring and analysis, which compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns.

[1002] 7. Detecting abnormal patterns and emotional states

[1003] The AI ​​system will mark a transaction as suspicious if it detects an abnormal trading pattern, and an emotional engine will make an additional determination as to whether a transaction is anomalous based on the user's emotional state.

[1004] 8. Suspicious Transaction Suspension

[1005] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[1006] 9. Send to human reviewers

[1007] The server sends the pending transaction data and associated sentiment data to human reviewers, who then scrutinize the data.

[1008] 10. Transaction Verification and Final Confirmation

[1009] Human reviewers consider the destination account, transaction time, transaction amount, and captured emotional state to make the final decision.

[1010] 11. Implementation of Final Decision

[1011] If the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1012] To give a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they are deemed legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[1013] This system will effectively prevent fraudulent transfers and irregular transactions in digital payment services, providing users with a service they can use with greater peace of mind.

[1014] The processing flow will be explained below.

[1015] Step 1:

[1016] The user inputs a remittance request using the terminal, specifies the recipient's account information and the amount to be remitted, and presses the remit button.

[1017] Step 2:

[1018] The terminal sends the generated transaction data (recipient account information, remittance amount, user ID, transaction time) to the server.

[1019] Step 3:

[1020] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it starts the emotion engine and starts acquiring the user's emotion data.

[1021] Step 4:

[1022] The device collects emotional data from users through built-in cameras, microphones, and other sensors, such as facial expressions, tone of voice, and heart rate.

[1023] Step 5:

[1024] The device sends the collected emotion data to the server, which then passes the data to the emotion engine.

[1025] Step 6:

[1026] The emotion engine analyzes the transmitted emotion data and determines the user's emotional state (e.g., anxious, nervous, relieved).

[1027] Step 7:

[1028] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with past transaction history and statistical data to detect abnormal patterns.

[1029] Step 8:

[1030] If the AI ​​system detects an abnormal transaction pattern, for example, if an amount that is excessively high compared to the amount normally sent, it will notify the server.

[1031] Step 9:

[1032] The server combines the results of the AI ​​system and the analysis results of the emotion engine to make a final decision on whether a transaction is suspicious. For example, if a transaction is abnormal and the user is in a tense state, the server will determine that the transaction is suspicious.

[1033] Step 10:

[1034] The server will temporarily suspend the suspicious transaction, set the transaction status to "Pending," and send a notification to the user stating, "The transaction is currently pending."

[1035] Step 11:

[1036] The server sends the pending transaction data and associated emotion data to a human reviewer, who is provided with the transaction details and emotional state.

[1037] Step 12:

[1038] Human reviewers will make a final decision based on the destination account, transaction time, transaction amount, and captured emotional data, taking into account, for example, if the transaction amount is significantly larger than normal or if the user appears to be in an unnatural state of stress.

[1039] Step 13:

[1040] If the reviewer approves the transaction, they send the approval result to the server, which updates the transaction status to "approved" and allows the transaction to proceed.

[1041] Step 14:

[1042] If the transaction is approved, the server proceeds with the actual transfer process. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[1043] Step 15:

[1044] If the reviewer decides to reject the transaction, the rejection is sent to the server, which updates the transaction status to "Cancelled" and cancels the transaction.

[1045] Step 16:

[1046] If the transaction is cancelled, the server sends a notification to the user's terminal that the transaction has been cancelled.

[1047] In this way, this system can combine and analyze transaction data and emotional data to detect and prevent suspicious transactions with high accuracy.

[1048] Example 2

[1049] 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."

[1050] In recent years, fraudulent remittances and abnormal transactions in digital payment services have been increasing, and conventional security measures are often inadequate to address these issues. New technologies are needed to detect fraudulent activity in advance and respond quickly. Furthermore, taking the user's emotional state into account is expected to enable more accurate detection of abnormal transactions, but achieving this requires advanced analytical technology and real-time processing. The present invention aims to solve these problems.

[1051] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving transaction data, means for monitoring the transaction data in real time, means for analyzing the transaction data and detecting abnormal transaction patterns, means for acquiring and analyzing user emotion data to identify the user's emotional state, means for temporarily suspending abnormal transactions based on the abnormal transactions and the emotion analysis results, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing results. This makes it possible to detect fraudulent transactions with high accuracy from both the transaction data and the emotion data, and to prevent fraud quickly and reliably.

[1052] "Transaction data" refers to data that includes detailed information such as the recipient account information specified by the user, the amount to be transferred, the user ID, and the time of the transaction.

[1053] "Real-time monitoring" means that as soon as transaction data is generated, its contents are immediately investigated and processing is carried out to detect any fraud or abnormalities.

[1054] "Abnormal trading patterns" refer to trading patterns that are clearly different from normal trading behavior or past trading history, or that are suspected of being fraudulent.

[1055] "Emotional data" refers to data regarding a user's emotional state, such as a user's facial expression, tone of voice, stress level, etc.

[1056] An "emotion engine" refers to software or algorithms that analyze emotional data and identify a user's emotional state.

[1057] "Temporary suspension" means that the abnormal transaction detected will not be immediately processed, but will be temporarily stopped for further inspection and confirmation.

[1058] "Reviewer" refers to a human professional who reviews transaction data and sentiment data to make a final determination regarding suspicious transactions.

[1059] "Final transaction processing" refers to the process of proceeding with or canceling a transfer based on the reviewer's determination of whether the transaction is fraudulent or not.

[1060] "Notification" refers to messages or information that inform a user about the status of a transaction or the final processing result.

[1061] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[1062] First, the user opens the application on the terminal and initiates a transaction. They enter a transfer request and specify the recipient's account information and the amount to be transferred. When the user presses the "Transfer" button, this information is generated as transaction data by the terminal. The terminal then encrypts the generated transaction data and sends it to the server via a secure network. The transaction data includes details such as the recipient's account information, the amount to be transferred, the user ID, and the time of the transaction.

[1063] The server receives transaction data sent from the terminal in real time and stores it in a database. The server also sends commands to the user's device to activate the camera, microphone, and other sensors to obtain the user's emotional data. The device collects the user's facial expressions, tone of voice, stress level, and other information in real time and generates emotional data. This emotional data is sent back to the server and analyzed by an emotion engine to identify the user's emotional state. For example, it determines states such as "anxiety" or "tension."

[1064] The server then passes the received transaction data to an AI system for real-time monitoring and analysis. The AI ​​system compares the transaction data with past transaction patterns and statistical data to detect abnormal transaction patterns. Based on the detected abnormal transactions and the results of sentiment analysis, the system temporarily suspends suspicious transactions and sets their status to "on hold." The server then sends a notification to the user stating, "The transaction is currently on hold."

[1065] The server then sends the pending transaction data and associated emotion data to a human reviewer who reviews the data and makes a final decision. The human reviewer reviews the destination account, transaction time, transaction amount, and captured emotion state. If necessary, the human reviewer can provide additional confirmation to the user.

[1066] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1067] This system effectively prevents fraudulent transfers and irregular transactions in digital payment services, allowing users to use the service with peace of mind.

[1068] As a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​system determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they determine that it is legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[1069] Prompt Sentence Examples

[1070] "Describe a scenario where a user is attempting to transfer a large amount of money. Detail how you would process it by analyzing sentiment data and detecting anomalous transactions."

[1071] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1072] Step 1:

[1073] The user opens the application on the terminal and enters a remittance request. The data entered includes the recipient's account information, the remittance amount, the user ID, and the transaction time. When the user presses the "Remit" button, this data is generated as transaction data by the terminal. The terminal then encrypts this transaction data and sends it to the server over a secure network.

[1074] Input: Remittance account information, remittance amount, user ID, transaction time

[1075] Output: Encrypted transaction data

[1076] Step 2:

[1077] The server receives the transaction data sent from the terminal and stores it in a database. The server also starts the emotion engine and prepares the user's device to collect emotion data.

[1078] Input: Encrypted transaction data

[1079] Output: Transaction data stored in a database

[1080] Step 3:

[1081] The server sends commands to the user's device to activate the camera, microphone, and other sensors, and the device collects the user's facial expressions, tone of voice, stress level, and other information in real time to generate emotional data.

[1082] Input: Commands to the user device

[1083] Output: User emotion data

[1084] Step 4:

[1085] Emotional data is sent from the user's device to the server, which then analyzes it with an emotion engine to identify the user's emotional state. For example, it determines whether the user is feeling anxious or nervous.

[1086] Input: Emotion data

[1087] Output: Parsed emotional state

[1088] Step 5:

[1089] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with historical transaction patterns and statistical data to detect abnormal transaction patterns.

[1090] Input: Transaction data

[1091] Output: Abnormal trading pattern detection results

[1092] Step 6:

[1093] If the AI ​​system detects an abnormal trading pattern, it will mark the transaction as suspicious, and if the emotion engine determines that the transaction is suspicious based on the user's emotional state, the transaction will also be marked as suspicious.

[1094] Input: Detected abnormal trading patterns, analyzed emotional state

[1095] Output: Suspicious transaction mark

[1096] Step 7:

[1097] The server temporarily holds the suspicious transaction and sets its status to "Pending." The server sends a notification to the user saying, "The transaction is currently pending."

[1098] Input: Mark suspicious transaction

[1099] Output: Pending transaction status, notification to user

[1100] Step 8:

[1101] The server sends the pending transaction data and associated sentiment data to human reviewers, who scrutinize the data and check for anomalies.

[1102] Input: Pending transaction data, associated sentiment data

[1103] Output: Send data to reviewers

[1104] Step 9:

[1105] A human reviewer will review the details of the destination account, transaction time, transaction amount, and captured emotional state. If necessary, the reviewer will ask the user for additional confirmation.

[1106] Input: Data to reviewer

[1107] Output: Final decision

[1108] Step 10:

[1109] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1110] Input: Final decision

[1111] Output: Transaction proceeds or is cancelled, and the user is notified

[1112] (Application example 2)

[1113] 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."

[1114] Increasing fraudulent remittances and abnormal transactions are becoming a major problem in modern digital payment services. This puts users at risk of fraudulent transactions and may undermine trust in payment services. Furthermore, conventional systems detect anomalies solely by analyzing transaction data, which means there is a risk of overlooking fraudulent transactions. Furthermore, accurate judgments are difficult because they do not take into account the user's emotional state. Therefore, a more accurate fraud detection system is needed.

[1115] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1116] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for acquiring user emotional data, means for analyzing the acquired emotional data to determine the user's emotional state, means for analyzing the monitored transaction data and temporarily suspending transactions determined to be anomalous based on anomalous transaction patterns and the user's emotional state, means for transmitting the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This makes it possible to improve the accuracy of detecting anomalous transactions and effectively reduce the risk of fraudulent remittances by taking the user's emotional state into consideration in addition to analyzing the transaction data.

[1117] "Transaction data" is information generated when a user transfers or makes a payment, and includes details such as the recipient's account information, the amount transferred, the user ID, and the time of the transaction.

[1118] An "abnormal transaction pattern" is a transaction that shows unnatural trends that differ from normal transactions and is an indicator of possible fraudulent transfers or other fraudulent activity.

[1119] "Emotion data" is data that reflects the user's emotional state, obtained by analyzing the user's facial expressions, tone of voice, stress level, etc.

[1120] The "emotional state" indicates the psychological state of the user, and refers to states such as impatience, tension, and relief.

[1121] "Temporary suspension" refers to the process of temporarily halting a transaction that has been determined to be abnormal and withholding a final decision.

[1122] A "human reviewer" is someone who manually reviews transactions and sentiment data held by the system and ultimately approves or rejects the transaction.

[1123] The system that realizes this application example consists of the following steps: First, a user launches a digital payment service application using a smartphone terminal. When the user enters a remittance request and specifies the recipient's account information and the remittance amount, the smartphone terminal generates transaction data and sends it to the server.

[1124] The server stores the received transaction data in a database. The transaction data includes details such as the recipient's account information, the remittance amount, the user ID, and the transaction time. At the same time, the user's emotion engine runs on the smartphone and collects the user's emotion data through the device's camera and microphone.

[1125] Emotion data includes the user's facial expressions, tone of voice, stress level, etc. The emotion engine analyzes the emotion data in real time to determine the user's emotional state. This emotion data is then sent to the server and integrated with the transaction data.

[1126] After integrating the above data, the server's AI system uses the received transaction data and emotional state data for real-time monitoring and analysis. The AI ​​system compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns, and then takes the emotional state into account to make a final anomalous transaction judgment. Transactions that are judged to be abnormal are temporarily suspended and sent to human reviewers for a final decision.

[1127] The reviewer verifies the validity of the transaction based on the pending transaction data and the user's emotional state data sent from the server. Based on the review results, the server performs the final transaction processing and notifies the user of the result.

[1128] The specific hardware and software used are as follows:

[1129] Smartphone device: camera, microphone, processing unit

[1130] Server: Database management system (MySQL, etc.), AI model (TensorFlow / PyTorch)

[1131] Emotion analysis engine: Voice analysis library (Librosa, etc.), face recognition library (OpenCV, etc.)

[1132] Monitoring system: System monitoring tools (Prometheus, etc.)

[1133] As a concrete example, consider a situation where a user attempts to transfer more than 100,000 yen. If the AI ​​detects an abnormality during the transfer process, such as the camera detecting a panicked expression or the microphone detecting a tense tone of voice, the transaction will be temporarily put on hold. The server then sends this data to a human reviewer, who will then approve or reject the transaction based on their final judgment.

[1134] Example prompt sentence:

[1135] "If a user attempts to send more than 100,000 yen using a smartphone app, consider a system that verifies the intention of the transfer. AI will monitor abnormal transactions and the user's facial expressions in real time to determine safety."

[1136] This invention improves the accuracy of detecting abnormal transactions by combining analysis of transaction data with data on the user's emotional state, making it possible to effectively reduce the risk of fraudulent remittances.

[1137] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1138] Step 1:

[1139] A user launches a digital payment service application on a smartphone terminal. The user enters a remittance request, specifying the recipient's account information and the amount to be sent. When the "Remit" button is pressed, the terminal generates transaction data.

[1140] Input: Recipient account information, remittance amount, user ID, transaction time

[1141] Output: Transaction data

[1142] Step 2:

[1143] The terminal transmits the generated transaction data to the server.

[1144] Input: Transaction data

[1145] Output: Transaction data sent to the server

[1146] Step 3:

[1147] The server stores the received transaction data in a database.

[1148] Input: Transaction data sent from the terminal

[1149] Output: Transaction data stored in a database

[1150] Step 4:

[1151] At the same time, the device's camera and microphone are used to collect the user's emotional data: the camera captures the user's facial expressions, and the microphone captures the tone of voice and stress level.

[1152] Input: Camera video, audio data

[1153] Output: Emotion data

[1154] Step 5:

[1155] The device analyzes the collected emotional data and determines the user's emotional state. Analysis is performed using a face recognition library (such as OpenCV) and a voice analysis library (such as Librosa).

[1156] Input: Emotion data

[1157] Output: Emotional state (e.g., anxious, nervous, relieved)

[1158] Step 6:

[1159] The device transmits the analyzed emotional state to the server.

[1160] Input: Parsed emotional state

[1161] Output: Emotional state sent to the server

[1162] Step 7:

[1163] The server combines the received transaction data with the emotional state data and passes it to the AI ​​system.

[1164] Input: Transaction data, emotional state data

[1165] Output: Integrated data (transaction data + emotional state data)

[1166] Step 8:

[1167] The AI ​​system monitors and analyzes the integrated data in real time, comparing it with previous trading patterns and statistical data to detect abnormal patterns, and then takes into account emotional states to make a final judgment on abnormal trading.

[1168] Inputs: Consolidated data, historical trading patterns, statistical data

[1169] Output: Abnormal transaction determination result

[1170] Step 9:

[1171] The server temporarily holds transactions that are identified as abnormal and sends the transaction data and emotional state data to a human reviewer.

[1172] Input: Abnormal transaction judgment results, transaction data, emotional state data

[1173] Output: Pending transactions, reviewer notifications

[1174] Step 10:

[1175] The reviewer evaluates the validity of the transaction based on the pending transaction data and emotional state data. If the transaction is deemed valid, the server proceeds with the transaction, but if there are any problems, the server cancels the transaction.

[1176] Input: Pending transaction data, emotional state data

[1177] Output: Review result (approved or rejected)

[1178] Step 11:

[1179] The server performs the final transaction processing based on the review results and notifies the user of the results.

[1180] Input: Review results

[1181] Output: Notification of final transaction result (transfer completion or transaction cancellation notification)

[1182] 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.

[1183] 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.

[1184] 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.

[1185] [Fourth embodiment]

[1186] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1187] 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.

[1188] 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).

[1189] 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.

[1190] 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.

[1191] 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).

[1192] 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.

[1193] 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.

[1194] 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.

[1195] 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.

[1196] 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.

[1197] 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.

[1198] 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."

[1199] This invention is a system for preventing fraudulent remittances in digital payment services. It uses AI to monitor transaction data in real time and detect abnormal transaction patterns. Furthermore, if a suspicious transaction is detected, the system temporarily suspends the transaction and leaves the final decision to a human reviewer, making a more accurate judgment.

[1200] A natural language description of what the program does

[1201] 1. Collection of transaction data

[1202] A user makes a remittance request from a terminal. For example, if a user attempts to remit 100,000 yen, transaction data (remittance destination information and transaction amount) is generated.

[1203] 2. Transmission of transaction data

[1204] The terminal transmits the generated transaction data to the server.

[1205] 3. Receiving Transaction Data

[1206] The server receives the transaction data sent from the terminal.

[1207] 4. Real-time monitoring and analysis

[1208] The server inputs the received transaction data into an AI algorithm, which analyzes the transaction data and detects abnormal patterns. For example, if a user who usually makes transactions of less than 10,000 yen suddenly transfers 100,000 yen, this would be an abnormal pattern.

[1209] 5. Detecting Abnormal Patterns

[1210] The AI ​​algorithm evaluates trading patterns against a user's past trading history and marks the transaction as suspicious if an abnormal trading pattern is detected.

[1211] 6. Suspicious Transaction Suspension

[1212] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[1213] 7. Send to human reviewers

[1214] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[1215] 8. Transaction Verification and Final Confirmation

[1216] Human reviewers check the recipient account, time of transaction, amount of transfer, etc. to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[1217] 9. Implementation of the Final Decision

[1218] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1219] To give a specific example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server and the AI ​​determines that it is an anomaly. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines that there are no problems, the transaction is finally completed. If there is a problem, the transaction is canceled and the user is notified.

[1220] This system effectively prevents fraudulent transfers via digital payment services, providing users with a service they can use with peace of mind.

[1221] The processing flow will be explained below.

[1222] Step 1:

[1223] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[1224] Step 2:

[1225] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[1226] Step 3:

[1227] The server receives the transaction data sent from the terminal and stores it in a database.

[1228] Step 4:

[1229] The server passes the transaction data to an AI system for real-time monitoring and analysis, which uses algorithms to analyze the transaction data and detect abnormal transaction patterns.

[1230] Step 5:

[1231] The AI ​​system compares transaction data with past transaction history and statistical data to assess whether there are any unusual patterns, for example detecting unusually high transfer amounts compared to the average of a user's past transactions.

[1232] Step 6:

[1233] If the AI ​​system detects an abnormal transaction pattern, the server marks the transaction as suspicious and puts it on hold, at which point the transaction's status is set to "pending."

[1234] Step 7:

[1235] The server sends a notification to the user's terminal that "The transaction is currently pending."

[1236] Step 8:

[1237] The server sends the pending transaction data to a human reviewer, who is provided with transaction details (destination account, transaction amount, user ID, etc.).

[1238] Step 9:

[1239] Human reviewers scrutinize transaction details, such as checking the trustworthiness of the recipient and whether similar transactions have been made before.

[1240] Step 10:

[1241] If the reviewer approves the transaction, the approval result is sent to the server, and the transaction status is updated to "Approved."

[1242] Step 11:

[1243] The server resumes the approved transaction and proceeds with the actual transfer. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[1244] Step 12:

[1245] If the reviewer decides to reject the transaction, they send the rejection result to the server, and the transaction status is set to "Cancelled" because the transaction was rejected.

[1246] Step 13:

[1247] The server cancels the transaction and sends a "transaction canceled" notification to the user's terminal.

[1248] This allows the system to detect abnormal transactions in real time and take appropriate action to prevent fraudulent transfers.

[1249] Example 1

[1250] 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."

[1251] Conventional digital payment systems have difficulty detecting sophisticated fraudulent remittances, preventing users from conducting transactions with peace of mind. Furthermore, they lacked mechanisms for detecting abnormal transaction patterns in real time and taking appropriate countermeasures. This increased the risk of fraudulent transactions, potentially resulting in significant losses for financial institutions and users. This invention aims to solve these issues by utilizing AI for advanced monitoring and analysis, effectively preventing fraudulent transactions.

[1252] 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.

[1253] In this invention, the server includes means for receiving transaction data sent by a user from a terminal, means for transmitting the received transaction data to the server via a secure communication means, means for storing the transaction data received by the server within the system, means for inputting the received transaction data into a generative AI model, means for monitoring and analyzing the transaction data in real time using the generative AI model, means for detecting anomalous transaction patterns from the analyzed transaction data, means for temporarily suspending the detected anomalous transaction and setting its status to "suspended," means for notifying the user of the suspended transaction data, means for transmitting the suspended transaction to a human reviewer, means for the reviewer to carefully examine the transaction content and make a judgment, means for final transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This enables real-time monitoring of transaction data and early detection of anomalies, thereby significantly reducing the risk of fraudulent transactions.

[1254] "Transaction data" is a collection of information related to remittances and payments made by users, including details such as the recipient account information and the remittance amount.

[1255] "Terminal" refers to the digital device used by a user, such as a smartphone or tablet, that provides the interface for making a remittance request.

[1256] A "server" is a computer system that receives, processes, and analyzes transaction data sent from a terminal.

[1257] A "generative AI model" is a model that uses artificial intelligence to analyze trading data and includes algorithms to detect anomalous trading patterns.

[1258] "Real-time monitoring" is the process of monitoring transaction data in real time and analyzing it immediately.

[1259] "Abnormal transaction patterns" refer to potentially fraudulent transaction behavior that deviates from normal transaction patterns and is detected by the AI ​​model.

[1260] "Suspicious transactions" are transactions that are marked as anomalous by the AI ​​model and require further review.

[1261] "Suspension" refers to a state in which a transaction in which an abnormality has been detected is temporarily halted and a final decision is pending.

[1262] A "reviewer" is a professional who examines pending transactions from a human perspective and makes a final decision.

[1263] "Final transaction processing" refers to the series of processes that approve or reject a transaction based on the reviewer's judgment.

[1264] "Notifications" are messages that inform users of the status or final outcome of a transaction.

[1265] The above definitions allow understanding of important terms related to this invention.

[1266] This invention is a system for preventing fraudulent remittances in digital payment services, specifically a system for real-time monitoring of transaction data and detection of abnormal patterns. The system includes major components such as users, terminals, and servers.

[1267] First, a user makes a remittance request using a device such as a smartphone or tablet. The user enters the recipient account information and the amount to be remitted, and transaction data is generated on the device. The generated transaction data includes the recipient information and the transaction amount.

[1268] The terminal encrypts the generated transaction data and then sends it to the server using a secure communication method (e.g., HTTPS). The server stores the transaction data received from the terminal in its system and verifies that the data is complete and accurate.

[1269] The server then inputs the received transaction data into a generative AI model, an algorithm that analyzes transaction data in real time against historical transaction data to detect anomalous transaction patterns. The AI ​​model identifies transactions that deviate from normal transaction patterns and marks transactions that are deemed potentially fraudulent as suspicious.

[1270] Any abnormal transaction detected will be temporarily suspended by the server and its status will be set to "Pending." At the same time, the server will send a notification to the user saying "The transaction is currently pending." This will let the user know that the transaction is pending and undergoing final confirmation.

[1271] The server then sends the pending transaction data to a human reviewer who examines the transaction details, checking the recipient account, time of day, amount, etc., to determine whether the transaction is legitimate. For example, if the recipient is newly registered, additional checks may be performed.

[1272] If the reviewer approves the transaction, the server will re-promote the transaction, or if the transaction is rejected, it will cancel the transaction and send a "Transaction Cancelled" notification to the user.

[1273] For example, if a user suddenly attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, where it is identified as an anomaly by the AI ​​model. The server then holds the transaction and sends it to a human reviewer. If the reviewer examines the transaction and determines there are no problems, the transaction is finally completed. If there are problems, the transaction is canceled and the user is notified.

[1274] An example of an input prompt for a generative AI model is:

[1275] Analyze the transaction data below to determine if there is any possibility of fraud.

[1276] Transaction Data:

[1277] User ID: 12345

[1278] Remittance destination: ABC Bank

[1279] Amount to be sent: 100,000 yen

[1280] Past trading patterns: Over 90% of transactions were under 10,000 yen

[1281] This transaction is unusual because it deviates significantly from the normal pattern.

[1282] This system effectively prevents fraudulent remittances in digital payment services, providing users with a service they can use with peace of mind. Also, through specific operational procedures, it is possible to understand in detail how this invention is implemented.

[1283] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1284] Step 1: Collect transaction data

[1285] The user makes a remittance request using the terminal. The user enters the remittance account information and the remittance amount, and transaction data is generated. This transaction data includes the remittance account information and the transaction amount.

[1286] (Input) The remittance destination information and transaction amount entered by the user into the terminal.

[1287] (Output) The generated transaction data.

[1288] Specific example of operation: The user opens the app, enters the recipient account information and the amount to send, and presses the "Send" button to generate transaction data.

[1289] Step 2: Sending transaction data

[1290] The terminal sends the generated transaction data to the server, which transmits the data using a secure communication method (e.g. HTTPS).

[1291] (Input) Transaction data generated by the terminal.

[1292] (Output) The encrypted transaction data sent to the server.

[1293] Specific operation example: The transaction data generated by the terminal is encrypted and sent to the server using the REST API.

[1294] Step 3: Receiving transaction data

[1295] The server receives the transaction data sent from the terminal, verifies that the data is complete and accurate, and stores the data in the system.

[1296] (Input) Encrypted transaction data sent from the terminal.

[1297] (Output) Transaction data stored in the system.

[1298] Specific operation example: The server saves the received data in a database and returns a reception confirmation response to the terminal.

[1299] Step 4: Real-time monitoring and analysis

[1300] The server inputs the received transaction data into a generative AI model, which analyzes the transaction data and detects anomalous patterns.

[1301] (Input) Transaction data stored on the server.

[1302] (Output) Trade valuation analyzed by the AI ​​model.

[1303] Specific example of operation: The server inputs transaction data into the generative AI model, and the AI ​​compares it with past transaction data and performs analysis.

[1304] Step 5: Detecting abnormal patterns

[1305] The generative AI model evaluates trading patterns against historical transaction data and marks transactions as suspicious if an abnormal trading pattern is detected.

[1306] (Input) Trade valuation analyzed by the AI ​​model.

[1307] (Output) Transactions marked as suspicious.

[1308] Specific example of operation: Transactions that the AI ​​determines to be abnormal are marked as "suspicious" and information about the suspicious transaction is recorded in a log.

[1309] Step 6: Suspicious Transaction Suspension

[1310] The server temporarily holds the suspicious transaction and sets its status to "Pending." The user is notified that "The transaction is currently pending."

[1311] (Input) Transactions marked as suspicious.

[1312] (Output) Transaction status set to "Pending" and user notification.

[1313] Example of specific behavior: The server updates the transaction status to "pending" and sends a pending notification to the user.

[1314] Step 7: Send to human reviewers

[1315] The server then sends the pending transaction data to a human reviewer, who examines the transaction and makes a decision.

[1316] (Input) "Pending" transaction data.

[1317] (Output) The transaction data sent to the reviewer.

[1318] Specific operation example: The server sends pending transaction data to the reviewer's system, and the reviewer receives the transaction data and starts reviewing it.

[1319] Step 8: Transaction verification and final confirmation

[1320] Human reviewers check the destination account, time of transaction, amount sent, etc. to determine whether the transaction is legitimate.

[1321] (Input) The transaction data received by the reviewer.

[1322] (Output) The transaction verification result.

[1323] Example of specific operation: A reviewer checks the remittance destination, transaction amount, etc., and asks the user for additional confirmation if necessary.

[1324] Step 9: Making a final decision

[1325] If a human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1326] (Input) Transaction verification result.

[1327] (Output) Final processing results and user notification.

[1328] Example of specific operation: A reviewer makes a decision on the system to approve or reject a transaction, and the server proceeds with approved transactions, cancels rejected transactions, and sends a final notification to the user.

[1329] (Application example 1)

[1330] 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."

[1331] The lack of a system to effectively prevent fraudulent transfers is a problem for existing digital payment services. In particular, it is difficult to monitor transaction data in real time, detect anomalies, temporarily suspend suspicious transactions, and promptly notify users and reviewers. This can lead to a lack of swift action when fraudulent transactions occur, which can result in a loss of user trust.

[1332] 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.

[1333] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for analyzing the monitored transaction data and detecting abnormal transaction patterns, means for temporarily suspending the detected abnormal transactions, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, means for notifying the user of the final processing result, and means for automatically notifying the user and the reviewer by email when an abnormal transaction is detected. This makes it possible to detect abnormal transactions in real time and respond to them quickly and efficiently.

[1334] "Transaction data" refers to information relating to transfers and payments made by users, including details such as the transfer destination, transfer amount, time, and transaction ID.

[1335] "Real-time monitoring means" refers to methods and systems for capturing transaction data in real time and analyzing it without delay.

[1336] "Means for detecting abnormal trading patterns" refers to AI algorithms or rule-based systems that identify transactions that are unusual or risky compared to normal trading history.

[1337] "Means for temporarily suspending abnormal transactions" refers to a system that has the ability to immediately stop and suspend a suspicious transaction when it is detected.

[1338] "Means of submission to human reviewer" refers to the communications methods and protocols used to submit details of a pending transaction to a human reviewer for final review.

[1339] "Means for finalizing transaction processing based on reviewer's judgment" refers to the system functionality for executing or canceling a transaction based on the reviewer's evaluation and decision.

[1340] "Means for notifying the user of the final processing result" refers to the notification method used to inform the user of the result of a transaction after it has been completed or canceled.

[1341] "Means for automatic email notification" refers to a system function that automatically sends emails to users and reviewers based on pre-set templates when an anomaly in transaction data is detected.

[1342] The system for realizing the present invention is configured based on a program including the following processing procedures.

[1343] First, a user makes a remittance request using a terminal. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This transaction data is sent to the server in real time.

[1344] The server receives the received transaction data and begins monitoring it in real time. It uses AI algorithms such as IsolationForest to detect abnormal transaction patterns and identify transactions that are unusual or high risk compared to normal transaction history (e.g., when transactions of several thousand yen are usually made, but a large amount is suddenly transferred).

[1345] Transactions that are identified as anomalous are temporarily put on hold and the transaction status is set to "Pending." Details of the pending transaction are sent to the user and human reviewer using an automated email notification feature. The user is notified that the transaction is currently on hold, and the reviewer is provided with details of the transaction.

[1346] Next, a reviewer will examine the submitted transaction data to determine whether or not the transaction is abnormal. This review will take into account the user's past transaction history, remittance destination information, and the time of the transaction. Based on the reviewer's decision, the final transaction processing will be carried out. If the transaction is approved, it will proceed; if it is rejected, it will be canceled.

[1347] The final transaction processing result will be notified to the user. Notifications will include messages such as "Transaction completed" or "Transaction canceled" using an automated email notification function.

[1348] This system mainly uses the following hardware and software:

[1349] Hardware: Devices such as smartphones

[1350] Software: Python, pandas (data processing), scikit-learn (AI modeling), smtplib (email sending)

[1351] For example, if a user normally sends a few thousand yen, but suddenly attempts to send 150,000 yen, the AI ​​will detect this suspicious transaction and put it on hold. The user and reviewers will be notified, and the reviewers will determine the legitimacy of the transaction.

[1352] An example prompt is:

[1353] Please provide a detailed review of the following transaction.

[1354] User ID: 1

[1355] Amount to be sent: 150,000 yen

[1356] Timestamp: 2023-10-10 16:00:00

[1357] New Recipient: True

[1358] The transaction is currently pending. You have been notified and can now decide whether to accept or cancel the transaction.

[1359] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1360] Step 1:

[1361] A user makes a remittance request using a smartphone. The remittance request includes the remittance amount, remittance destination information, and a timestamp. This generates transaction data. The inputs are the "remittance amount," "remittance destination information," and "timestamp," and the output is "transaction data."

[1362] Step 2:

[1363] The terminal sends the generated transaction data to the server. Based on the transaction data, the server receives the data. The input here is "transaction data" and the output is "transaction data received by server."

[1364] Step 3:

[1365] The server begins monitoring transaction data in real time. The received transaction data is input into an AI algorithm such as IsolationForest, which analyzes it to detect abnormal transaction patterns. The input is "transaction data" and the output is "detection results of abnormal patterns." Specifically, each field of the transaction data (such as the amount sent and the recency of the recipient) is applied to the AI ​​algorithm.

[1366] Step 4:

[1367] Transactions that are determined to be abnormal are temporarily suspended and the transaction status is set to "pending." This suspension process causes the server to mark the abnormal transaction as a flag and handle it specially in subsequent processes. The input is the "detection result of an abnormal transaction," and the output is the "pending transaction data."

[1368] Step 5:

[1369] The server automatically sends emails with details of the released transaction to the user and reviewer. Specifically, emails are sent using smtplib. The input is "pending transaction data" and the output is "sent email."

[1370] Step 6:

[1371] The reviewer scrutinizes the transaction data received by email and determines whether or not the transaction is anomalous. The input here is the "transaction data received by email" and the output is the "review results." Specifically, the reviewer determines the validity of the anomalous transaction based on the transaction history, remittance destination information, transaction time period, etc.

[1372] Step 7:

[1373] Based on the reviewer's decision, the server performs the final transaction processing. If the transaction is approved, the transaction data is processed as usual. If it is rejected, the transaction is canceled. The input is the "review result" and the output is the "final transaction processing result."

[1374] Step 8:

[1375] The server notifies the user of the final processing result. This notification is sent via automatic email using smtplib. The input is the "final transaction processing result" and the output is "notification to the user." Specifically, an email is sent to the user stating the status of the transaction, whether it was approved or canceled.

[1376] 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.

[1377] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[1378] A natural language description of what the program does

[1379] 1. Collection of transaction data

[1380] The user inputs a remittance request into the terminal, specifying the recipient's account information and the amount to be remitted. When the user presses the "Remit" button, this information is generated as transaction data by the terminal.

[1381] 2. Transmission of transaction data

[1382] The terminal sends the generated transaction data to the server, which includes details such as the recipient account information, the amount to be transferred, the user ID, and the time of the transaction.

[1383] 3. Receiving Transaction Data

[1384] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it also starts the user's emotion engine.

[1385] 4. Acquiring Emotion Data

[1386] It captures emotional data about users through cameras, microphones, and other sensors built into their devices, such as facial expressions, tone of voice, and stress levels.

[1387] 5. Emotion Data Analysis

[1388] The emotion engine analyzes the acquired emotion data and determines the user's emotional state (for example, impatience, tension, relief).

[1389] 6. Real-time monitoring and analysis

[1390] The server then passes the received transaction data to the AI ​​system for real-time monitoring and analysis, which compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns.

[1391] 7. Detecting abnormal patterns and emotional states

[1392] The AI ​​system will mark a transaction as suspicious if it detects an abnormal trading pattern, and an emotional engine will make an additional determination as to whether a transaction is anomalous based on the user's emotional state.

[1393] 8. Suspicious Transaction Suspension

[1394] The server temporarily holds the suspicious transaction and sets its status to "Pending." At this point, the user is notified that "The transaction is currently pending."

[1395] 9. Send to human reviewers

[1396] The server sends the pending transaction data and associated sentiment data to human reviewers, who then scrutinize the data.

[1397] 10. Transaction Verification and Final Confirmation

[1398] Human reviewers consider the destination account, transaction time, transaction amount, and captured emotional state to make the final decision.

[1399] 11. Implementation of Final Decision

[1400] If the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1401] To give a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they are deemed legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[1402] This system will effectively prevent fraudulent transfers and irregular transactions in digital payment services, providing users with a service they can use with greater peace of mind.

[1403] The processing flow will be explained below.

[1404] Step 1:

[1405] The user inputs a remittance request using the terminal, specifies the recipient's account information and the amount to be remitted, and presses the remit button.

[1406] Step 2:

[1407] The terminal sends the generated transaction data (recipient account information, remittance amount, user ID, transaction time) to the server.

[1408] Step 3:

[1409] The server receives the transaction data sent from the terminal and stores it in a database. At the same time, it starts the emotion engine and starts acquiring the user's emotion data.

[1410] Step 4:

[1411] The device collects emotional data from users through built-in cameras, microphones, and other sensors, such as facial expressions, tone of voice, and heart rate.

[1412] Step 5:

[1413] The device sends the collected emotion data to the server, which then passes the data to the emotion engine.

[1414] Step 6:

[1415] The emotion engine analyzes the transmitted emotion data and determines the user's emotional state (e.g., anxious, nervous, relieved).

[1416] Step 7:

[1417] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with past transaction history and statistical data to detect abnormal patterns.

[1418] Step 8:

[1419] If the AI ​​system detects an abnormal transaction pattern, for example, if an amount that is excessively high compared to the amount normally sent, it will notify the server.

[1420] Step 9:

[1421] The server combines the results of the AI ​​system and the analysis results of the emotion engine to make a final decision on whether a transaction is suspicious. For example, if a transaction is abnormal and the user is in a tense state, the server will determine that the transaction is suspicious.

[1422] Step 10:

[1423] The server will temporarily suspend the suspicious transaction, set the transaction status to "Pending," and send a notification to the user stating, "The transaction is currently pending."

[1424] Step 11:

[1425] The server sends the pending transaction data and associated emotion data to a human reviewer, who is provided with the transaction details and emotional state.

[1426] Step 12:

[1427] Human reviewers will make a final decision based on the destination account, transaction time, transaction amount, and captured emotional data, taking into account, for example, if the transaction amount is significantly larger than normal or if the user appears to be in an unnatural state of stress.

[1428] Step 13:

[1429] If the reviewer approves the transaction, they send the approval result to the server, which updates the transaction status to "approved" and allows the transaction to proceed.

[1430] Step 14:

[1431] If the transaction is approved, the server proceeds with the actual transfer process. Once the transfer is complete, the server sends a notification to the user that the transaction has been completed.

[1432] Step 15:

[1433] If the reviewer decides to reject the transaction, the rejection is sent to the server, which updates the transaction status to "Cancelled" and cancels the transaction.

[1434] Step 16:

[1435] If the transaction is cancelled, the server sends a notification to the user's terminal that the transaction has been cancelled.

[1436] In this way, this system can combine and analyze transaction data and emotional data to detect and prevent suspicious transactions with high accuracy.

[1437] Example 2

[1438] 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."

[1439] In recent years, fraudulent remittances and abnormal transactions in digital payment services have been increasing, and conventional security measures are often inadequate to address these issues. New technologies are needed to detect fraudulent activity in advance and respond quickly. Furthermore, taking the user's emotional state into account is expected to enable more accurate detection of abnormal transactions, but achieving this requires advanced analytical technology and real-time processing. The present invention aims to solve these problems.

[1440] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving transaction data, means for monitoring the transaction data in real time, means for analyzing the transaction data and detecting abnormal transaction patterns, means for acquiring and analyzing user emotion data to identify the user's emotional state, means for temporarily suspending abnormal transactions based on the abnormal transactions and the emotion analysis results, means for sending the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing results. This makes it possible to detect fraudulent transactions with high accuracy from both the transaction data and the emotion data, and to prevent fraud quickly and reliably.

[1441] "Transaction data" refers to data that includes detailed information such as the recipient account information specified by the user, the amount to be transferred, the user ID, and the time of the transaction.

[1442] "Real-time monitoring" means that as soon as transaction data is generated, its contents are immediately investigated and processing is carried out to detect any fraud or abnormalities.

[1443] "Abnormal trading patterns" refer to trading patterns that are clearly different from normal trading behavior or past trading history, or that are suspected of being fraudulent.

[1444] "Emotional data" refers to data regarding a user's emotional state, such as a user's facial expression, tone of voice, stress level, etc.

[1445] An "emotion engine" refers to software or algorithms that analyze emotional data and identify a user's emotional state.

[1446] "Temporary suspension" means that the abnormal transaction detected will not be immediately processed, but will be temporarily stopped for further inspection and confirmation.

[1447] "Reviewer" refers to a human professional who reviews transaction data and sentiment data to make a final determination regarding suspicious transactions.

[1448] "Final transaction processing" refers to the process of proceeding with or canceling a transfer based on the reviewer's determination of whether the transaction is fraudulent or not.

[1449] "Notification" refers to messages or information that inform a user about the status of a transaction or the final processing result.

[1450] This invention is a system for preventing fraudulent remittances in digital payment services, and utilizes AI to monitor transaction data in real time and detect abnormal transaction patterns and user emotions to make more accurate judgments. This system not only analyzes transaction data, but also detects the user's emotional state and uses that information to determine whether a transaction is abnormal.

[1451] First, the user opens the application on the terminal and initiates a transaction. They enter a transfer request and specify the recipient's account information and the amount to be transferred. When the user presses the "Transfer" button, this information is generated as transaction data by the terminal. The terminal then encrypts the generated transaction data and sends it to the server via a secure network. The transaction data includes details such as the recipient's account information, the amount to be transferred, the user ID, and the time of the transaction.

[1452] The server receives transaction data sent from the terminal in real time and stores it in a database. The server also sends commands to the user's device to activate the camera, microphone, and other sensors to obtain the user's emotional data. The device collects the user's facial expressions, tone of voice, stress level, and other information in real time and generates emotional data. This emotional data is sent back to the server and analyzed by an emotion engine to identify the user's emotional state. For example, it determines states such as "anxiety" or "tension."

[1453] The server then passes the received transaction data to an AI system for real-time monitoring and analysis. The AI ​​system compares the transaction data with past transaction patterns and statistical data to detect abnormal transaction patterns. Based on the detected abnormal transactions and the results of sentiment analysis, the system temporarily suspends suspicious transactions and sets their status to "on hold." The server then sends a notification to the user stating, "The transaction is currently on hold."

[1454] The server then sends the pending transaction data and associated emotion data to a human reviewer who reviews the data and makes a final decision. The human reviewer reviews the destination account, transaction time, transaction amount, and captured emotion state. If necessary, the human reviewer can provide additional confirmation to the user.

[1455] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1456] This system effectively prevents fraudulent transfers and irregular transactions in digital payment services, allowing users to use the service with peace of mind.

[1457] As a specific example, when a user attempts to transfer a large amount of money using a terminal, the transaction data is sent to the server, and emotional data is collected in parallel. If the AI ​​system determines that something is abnormal and the emotion engine detects that the user is anxious or nervous, the transaction is put on hold and the final decision is left to a human reviewer. The reviewer checks the transaction details and emotional state, and if they determine that it is legitimate, the transaction is completed. If a problem is determined, the transaction is canceled and the user is notified.

[1458] Prompt Sentence Examples

[1459] "Describe a scenario where a user is attempting to transfer a large amount of money. Detail how you would process it by analyzing sentiment data and detecting anomalous transactions."

[1460] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1461] Step 1:

[1462] The user opens the application on the terminal and enters a remittance request. The data entered includes the recipient's account information, the remittance amount, the user ID, and the transaction time. When the user presses the "Remit" button, this data is generated as transaction data by the terminal. The terminal then encrypts this transaction data and sends it to the server over a secure network.

[1463] Input: Remittance account information, remittance amount, user ID, transaction time

[1464] Output: Encrypted transaction data

[1465] Step 2:

[1466] The server receives the transaction data sent from the terminal and stores it in a database. The server also starts the emotion engine and prepares the user's device to collect emotion data.

[1467] Input: Encrypted transaction data

[1468] Output: Transaction data stored in a database

[1469] Step 3:

[1470] The server sends commands to the user's device to activate the camera, microphone, and other sensors, and the device collects the user's facial expressions, tone of voice, stress level, and other information in real time to generate emotional data.

[1471] Input: Commands to the user device

[1472] Output: User emotion data

[1473] Step 4:

[1474] Emotional data is sent from the user's device to the server, which then analyzes it with an emotion engine to identify the user's emotional state. For example, it determines whether the user is feeling anxious or nervous.

[1475] Input: Emotion data

[1476] Output: Parsed emotional state

[1477] Step 5:

[1478] The server passes the transaction data to an AI system for real-time monitoring and analysis, which compares the transaction data with historical transaction patterns and statistical data to detect abnormal transaction patterns.

[1479] Input: Transaction data

[1480] Output: Abnormal trading pattern detection results

[1481] Step 6:

[1482] If the AI ​​system detects an abnormal trading pattern, it will mark the transaction as suspicious, and if the emotion engine determines that the transaction is suspicious based on the user's emotional state, the transaction will also be marked as suspicious.

[1483] Input: Detected abnormal trading patterns, analyzed emotional state

[1484] Output: Suspicious transaction mark

[1485] Step 7:

[1486] The server temporarily holds the suspicious transaction and sets its status to "Pending." The server sends a notification to the user saying, "The transaction is currently pending."

[1487] Input: Mark suspicious transaction

[1488] Output: Pending transaction status, notification to user

[1489] Step 8:

[1490] The server sends the pending transaction data and associated sentiment data to human reviewers, who scrutinize the data and check for anomalies.

[1491] Input: Pending transaction data, associated sentiment data

[1492] Output: Send data to reviewers

[1493] Step 9:

[1494] A human reviewer will review the details of the destination account, transaction time, transaction amount, and captured emotional state. If necessary, the reviewer will ask the user for additional confirmation.

[1495] Input: Data to reviewer

[1496] Output: Final decision

[1497] Step 10:

[1498] Finally, if the human reviewer approves the transaction, the server allows it to proceed. If the transaction is rejected, the server cancels the transaction and notifies the user that the transaction has been canceled.

[1499] Input: Final decision

[1500] Output: Transaction proceeds or is cancelled, and the user is notified

[1501] (Application example 2)

[1502] 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."

[1503] Increasing fraudulent remittances and abnormal transactions are becoming a major problem in modern digital payment services. This puts users at risk of fraudulent transactions and may undermine trust in payment services. Furthermore, conventional systems detect anomalies solely by analyzing transaction data, which means there is a risk of overlooking fraudulent transactions. Furthermore, accurate judgments are difficult because they do not take into account the user's emotional state. Therefore, a more accurate fraud detection system is needed.

[1504] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1505] In this invention, the server includes means for receiving transaction data, means for monitoring the received transaction data in real time, means for acquiring user emotional data, means for analyzing the acquired emotional data to determine the user's emotional state, means for analyzing the monitored transaction data and temporarily suspending transactions determined to be anomalous based on anomalous transaction patterns and the user's emotional state, means for transmitting the suspended transactions to a human reviewer, means for finalizing the transaction processing based on the reviewer's judgment, and means for notifying the user of the final processing result. This makes it possible to improve the accuracy of detecting anomalous transactions and effectively reduce the risk of fraudulent remittances by taking the user's emotional state into consideration in addition to analyzing the transaction data.

[1506] "Transaction data" is information generated when a user transfers or makes a payment, and includes details such as the recipient's account information, the amount transferred, the user ID, and the time of the transaction.

[1507] An "abnormal transaction pattern" is a transaction that shows unnatural trends that differ from normal transactions and is an indicator of possible fraudulent transfers or other fraudulent activity.

[1508] "Emotion data" is data that reflects the user's emotional state, obtained by analyzing the user's facial expressions, tone of voice, stress level, etc.

[1509] The "emotional state" indicates the psychological state of the user, and refers to states such as impatience, tension, and relief.

[1510] "Temporary suspension" refers to the process of temporarily halting a transaction that has been determined to be abnormal and withholding a final decision.

[1511] A "human reviewer" is someone who manually reviews transactions and sentiment data held by the system and ultimately approves or rejects the transaction.

[1512] The system that realizes this application example consists of the following steps: First, a user launches a digital payment service application using a smartphone terminal. When the user enters a remittance request and specifies the recipient's account information and the remittance amount, the smartphone terminal generates transaction data and sends it to the server.

[1513] The server stores the received transaction data in a database. The transaction data includes details such as the recipient's account information, the remittance amount, the user ID, and the transaction time. At the same time, the user's emotion engine runs on the smartphone and collects the user's emotion data through the device's camera and microphone.

[1514] Emotion data includes the user's facial expressions, tone of voice, stress level, etc. The emotion engine analyzes the emotion data in real time to determine the user's emotional state. This emotion data is then sent to the server and integrated with the transaction data.

[1515] After integrating the above data, the server's AI system uses the received transaction data and emotional state data for real-time monitoring and analysis. The AI ​​system compares the transaction data with previous transaction patterns and statistical data to detect abnormal patterns, and then takes the emotional state into account to make a final anomalous transaction judgment. Transactions that are judged to be abnormal are temporarily suspended and sent to human reviewers for a final decision.

[1516] The reviewer verifies the validity of the transaction based on the pending transaction data and the user's emotional state data sent from the server. Based on the review results, the server performs the final transaction processing and notifies the user of the result.

[1517] The specific hardware and software used are as follows:

[1518] Smartphone device: camera, microphone, processing unit

[1519] Server: Database management system (MySQL, etc.), AI model (TensorFlow / PyTorch)

[1520] Emotion analysis engine: Voice analysis library (Librosa, etc.), face recognition library (OpenCV, etc.)

[1521] Monitoring system: System monitoring tools (Prometheus, etc.)

[1522] As a concrete example, consider a situation where a user attempts to transfer more than 100,000 yen. If the AI ​​detects an abnormality during the transfer process, such as the camera detecting a panicked expression or the microphone detecting a tense tone of voice, the transaction will be temporarily put on hold. The server then sends this data to a human reviewer, who will then approve or reject the transaction based on their final judgment.

[1523] Example prompt sentence:

[1524] "If a user attempts to send more than 100,000 yen using a smartphone app, consider a system that verifies the intention of the transfer. AI will monitor abnormal transactions and the user's facial expressions in real time to determine safety."

[1525] This invention improves the accuracy of detecting abnormal transactions by combining analysis of transaction data with data on the user's emotional state, making it possible to effectively reduce the risk of fraudulent remittances.

[1526] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1527] Step 1:

[1528] A user launches a digital payment service application on a smartphone terminal. The user enters a remittance request, specifying the recipient's account information and the amount to be sent. When the "Remit" button is pressed, the terminal generates transaction data.

[1529] Input: Recipient account information, remittance amount, user ID, transaction time

[1530] Output: Transaction data

[1531] Step 2:

[1532] The terminal transmits the generated transaction data to the server.

[1533] Input: Transaction data

[1534] Output: Transaction data sent to the server

[1535] Step 3:

[1536] The server stores the received transaction data in a database.

[1537] Input: Transaction data sent from the terminal

[1538] Output: Transaction data stored in a database

[1539] Step 4:

[1540] At the same time, the device's camera and microphone are used to collect the user's emotional data: the camera captures the user's facial expressions, and the microphone captures the tone of voice and stress level.

[1541] Input: Camera video, audio data

[1542] Output: Emotion data

[1543] Step 5:

[1544] The device analyzes the collected emotional data and determines the user's emotional state. Analysis is performed using a face recognition library (such as OpenCV) and a voice analysis library (such as Librosa).

[1545] Input: Emotion data

[1546] Output: Emotional state (e.g., anxious, nervous, relieved)

[1547] Step 6:

[1548] The device transmits the analyzed emotional state to the server.

[1549] Input: Parsed emotional state

[1550] Output: Emotional state sent to the server

[1551] Step 7:

[1552] The server combines the received transaction data with the emotional state data and passes it to the AI ​​system.

[1553] Input: Transaction data, emotional state data

[1554] Output: Integrated data (transaction data + emotional state data)

[1555] Step 8:

[1556] The AI ​​system monitors and analyzes the integrated data in real time, comparing it with previous trading patterns and statistical data to detect abnormal patterns, and then takes into account emotional states to make a final judgment on abnormal trading.

[1557] Inputs: Consolidated data, historical trading patterns, statistical data

[1558] Output: Abnormal transaction determination result

[1559] Step 9:

[1560] The server temporarily holds transactions that are identified as abnormal and sends the transaction data and emotional state data to a human reviewer.

[1561] Input: Abnormal transaction judgment results, transaction data, emotional state data

[1562] Output: Pending transactions, reviewer notifications

[1563] Step 10:

[1564] The reviewer evaluates the validity of the transaction based on the pending transaction data and emotional state data. If the transaction is deemed valid, the server proceeds with the transaction, but if there are any problems, the server cancels the transaction.

[1565] Input: Pending transaction data, emotional state data

[1566] Output: Review result (approved or rejected)

[1567] Step 11:

[1568] The server performs the final transaction processing based on the review results and notifies the user of the results.

[1569] Input: Review results

[1570] Output: Notification of final transaction result (transfer completion or transaction cancellation notification)

[1571] 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.

[1572] 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.

[1573] 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.

[1574] 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.

[1575] 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.

[1576] 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.

[1577] 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).

[1578] 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.

[1579] 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."

[1580] 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.

[1581] 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).

[1582] 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.

[1583] 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.

[1584] 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.

[1585] 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.

[1586] 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.

[1587] 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.

[1588] 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.

[1589] 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.

[1590] 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.

[1591] 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.

[1592] The following is further disclosed regarding the above embodiment.

[1593] (Claim 1)

[1594] means for receiving transaction data;

[1595] means for monitoring received transaction data in real time;

[1596] means for analyzing the monitored transaction data to detect anomalous transaction patterns;

[1597] A means for temporarily suspending detected abnormal transactions;

[1598] a means for sending the pending transaction to a human reviewer;

[1599] A means for finalizing the transaction based on the reviewer's decision;

[1600] means for notifying the user of the final processing result;

[1601] A system including:

[1602] (Claim 2)

[1603] The system according to claim 1, characterized in that an algorithm is used to detect abnormal patterns by comparing the reserved transaction data with past transaction data.

[1604] (Claim 3)

[1605] The system according to claim 1, further comprising means for verifying a user's past transaction history, a transfer destination account, and a time period for the transaction.

[1606] "Example 1"

[1607] (Claim 1)

[1608] means for receiving transaction data transmitted by a user from a terminal;

[1609] means for transmitting the received transaction data to a server via a secure communication means;

[1610] means for storing the transaction data received by the server within the system;

[1611] a means for inputting the received transaction data into a generative AI model;

[1612] A means to monitor and analyze transaction data in real time using generative AI models; and

[1613] means for detecting anomalous trading patterns from the analyzed trading data;

[1614] a means for temporarily suspending the detected abnormal transaction and setting its status to "suspended";

[1615] means for notifying a user of the withheld transaction data;

[1616] a means for sending the pending transaction to a human reviewer;

[1617] A means for reviewers to examine and judge the transaction details;

[1618] A means for finalizing the transaction based on the reviewer's decision;

[1619] means for notifying the user of the final processing result;

[1620] A system including:

[1621] (Claim 2)

[1622] The system described in claim 1 uses a generative AI model that compares the reserved transaction data with past transaction data to detect abnormal patterns.

[1623] (Claim 3)

[1624] The system according to claim 1, further comprising means for verifying a user's past transaction history, a transfer destination account, and a time period for the transaction.

[1625] "Application Example 1"

[1626] (Claim 1)

[1627] means for receiving transaction data;

[1628] means for monitoring received transaction data in real time;

[1629] means for analyzing the monitored transaction data to detect anomalous transaction patterns;

[1630] A means for temporarily suspending detected abnormal transactions;

[1631] a means for sending the pending transaction to a human reviewer;

[1632] A means for finalizing the transaction based on the reviewer's decision;

[1633] a means for notifying the user of the final processing result;

[1634] A means of automatically notifying users and reviewers by email when an abnormal transaction is detected;

[1635] A system including:

[1636] (Claim 2)

[1637] The system of claim 1 uses an algorithm to compare the reserved transaction data with past transaction data to detect abnormal patterns.

[1638] (Claim 3)

[1639] The system of claim 1, further comprising means for verifying a user's past transaction history, remittance destination information, and transaction time period.

[1640] "Example 2: Combining Emotion Engines"

[1641] (Claim 1)

[1642] means for receiving transaction data;

[1643] means for monitoring received transaction data in real time;

[1644] means for analyzing the monitored transaction data to detect anomalous transaction patterns;

[1645] means for acquiring and analyzing user emotional data to identify the user's emotional state;

[1646] a means for temporarily suspending abnormal transactions based on the detected abnormal transactions and sentiment analysis results;

[1647] a means for sending the pending transaction to a human reviewer;

[1648] A means for finalizing the transaction based on the reviewer's decision;

[1649] means for notifying the user of the final processing result;

[1650] A system including:

[1651] (Claim 2)

[1652] The system according to claim 1, characterized in that an algorithm is used to detect abnormal patterns by comparing the reserved transaction data with past transaction data.

[1653] (Claim 3)

[1654] 10. The system of claim 1, further comprising means for verifying a user's past transaction history, a destination account, a time period during which the transaction occurred, and a captured emotional state.

[1655] "Application example 2 when combining emotion engines"

[1656] (Claim 1)

[1657] means for receiving transaction data;

[1658] means for monitoring received transaction data in real time;

[1659] means for analyzing the monitored transaction data to detect anomalous transaction patterns;

[1660] A means for acquiring user emotion data;

[1661] means for analyzing the acquired emotion data to determine the user's emotional state;

[1662] means for temporarily suspending transactions determined to be abnormal based on the detected abnormal transactions and emotional state;

[1663] a means for sending the pending transaction to a human reviewer;

[1664] A means for finalizing the transaction based on the reviewer's decision;

[1665] means for notifying the user of the final processing result;

[1666] A system including:

[1667] (Claim 2)

[1668] The system according to claim 1, characterized in that an algorithm is used to detect abnormal patterns by comparing the reserved transaction data with past transaction data.

[1669] (Claim 3)

[1670] 10. The system of claim 1, further comprising means for verifying a user's past transaction history, destination account, time of transaction, and emotional state. [Explanation of symbols]

[1671] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving transaction data; means for monitoring received transaction data in real time; means for analyzing the monitored transaction data to detect anomalous transaction patterns; A means for temporarily suspending detected abnormal transactions; a means for sending the pending transaction to a human reviewer; A means for finalizing the transaction based on the reviewer's decision; means for notifying the user of the final processing result; A system including:

2. The system according to claim 1, wherein an algorithm is used to detect abnormal patterns by comparing the reserved transaction data with past transaction data.

3. 2. The system according to claim 1, further comprising means for verifying a user's past transaction history, a transfer destination account, and a time period for the transaction.

Images