Welfare assistance optimization system

Abstract

The spread of cashless payments has made it difficult to accurately ascertain the income of public assistance recipients through traditional bank account inquiries alone, leading to fraudulent claims. Furthermore, the aging of those responsible for monitoring recipients and a shortage of personnel have resulted in insufficient verification of the recipients' well-being. [Solution] The system includes a data collection means that integrates transaction history from bank accounts, electronic money, and QR code payments by linking with the interfaces of multiple payment services; a determination means that enables quick reporting via a smartphone app when an undeclared deposit is detected; and a data fixing means that records all declared information in an unalterable format including a hash value. Furthermore, it includes an evaluation means that detects discontinuation in transaction history and performs safety checks through automatic phone calls by an artificial intelligence voice agent, thereby supporting the work of the monitoring staff.

Description

Technical Field

[0001] The present invention relates to an information processing system that helps prevent fraud in public assistance systems, reduces the reporting burden on recipients, reduces the workload of welfare workers (hereinafter including "caseworkers") and community watchers (hereinafter including "neighborhood commissioners"), and supports the proper use of benefits.

Background Art

[0002] In the Japanese public assistance system, recipients are obliged to report their income, and administrative agencies have the authority to investigate the income and asset status of recipients. Conventionally, administrative agencies have grasped the income of recipients by inquiring about bank accounts. However, the said bank account inquiry is a manual procedure of mailing a paper inquiry form to a financial institution and waiting for a reply, which takes several weeks for each case.

[0003] However, due to the rapid spread of cashless payments in recent years, the number of cases where salaries, rewards, personal transfers, etc. are made directly through electronic money or two-dimensional code payment services without going through a bank account has been increasing.

[0004] As a result, it has become difficult to accurately grasp the income of recipients only by the conventional bank account inquiry, and the traceability of income is lacking. This has led to fraud in which recipients intentionally conceal their income or mistakes due to forgetting to report.

[0005] Furthermore, recipients are obliged to handwrite a monthly income report form regardless of their income and submit it to the welfare office together with a salary slip and a copy of the passbook. This paper-based reporting procedure is a great burden on recipients and is also a cause of reporting delays and entry errors.

[0006] Furthermore, community watch personnel (corresponding to terminal 140 in Figure 1, including "community welfare commissioners"), who are responsible for monitoring residents' living conditions and ensuring their well-being, are reaching their limits due to an aging workforce and a shortage of personnel. The average age of community welfare commissioners exceeds 66, and the staffing rate continues to decline, with some areas having one commissioner responsible for more than 130 households without compensation. In this situation, regular visits by community welfare commissioners alone are insufficient to adequately confirm the well-being of recipients such as elderly people living alone and people with disabilities.

[0007] In addition, welfare office caseworkers (corresponding to approvers in Figure 6) are responsible for 80 to over 100 households each, and are overwhelmed with tasks such as home visits, verification of income declarations, and processing of benefit decisions. Caseworkers are generally transferred every three years, but information on assigned households relies on verbal and written handover, which creates a problem where there is a gap in understanding the recipients' situations during the handover. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Patent No. 5692586 [Patent Document 2] Patent No. 5503795 [Overview of the project] [Problems that the invention aims to solve]

[0009] The present invention aims to ensure traceability of income and expenditure across multiple payment methods, reduce the burden of monthly paper declarations for recipients by replacing them with a 24-second smartphone operation while preventing fraudulent claims, reduce the burden of physical visits by welfare commissioners facing an aging workforce and a shortage of personnel through automated welfare checks, and reduce the administrative burden on caseworkers through priority display and automated report generation. [Means for solving the problem]

[0010] To solve the above problems, the fraud prevention system for public assistance according to the present invention is connected via a network to an information terminal used by public assistance recipients, payment servers managed by multiple payment service providers, and welfare management servers managed by administrative agencies.

[0011] This system includes data collection means for integrating transaction history through the interfaces of multiple payment services, determination means for detecting undeclared deposits and sending notifications to the recipient's information terminal, and data fixing means for recording response content in an unalterable format. [Effects of the Invention]

[0012] The present invention provides the following benefits. First, it ensures traceability of income across bank accounts, electronic money, and QR code payments. Second, recipients can complete their income declarations in under 24 seconds on a smartphone app instead of having to create and submit handwritten monthly income declaration forms (see Figure 2). Third, welfare commissioners can ascertain the well-being of all assigned households without physically visiting them, through automated phone calls by an artificial intelligence voice agent (see Figure 10) and monitoring of utility usage by smart meters (see Figure 11). Fourth, caseworkers can prioritize households with the highest urgency based on priority displays on the dashboard (classified by green, yellow, and red in Figure 7), and reduce administrative processing time through an automatic monthly report generation function (see Figure 7). [Brief explanation of the drawing]

[0013] [Figure 1] This is a diagram showing the overall system configuration of the present invention. It shows the connection relationships between the recipient's information terminal 110, payment server 120, welfare management server 130, civil affairs commissioner terminal 140 (indicated as "Civil Affairs Commissioner Terminal" in the diagram), server 150, and evidence storage room 160. [Figure 2] This is an income reporting flowchart. It shows that the entire process, from deposit detection (step 200) to hash-verified recording (step 260), can be completed within 24 seconds. [Figure 3]This is a dual monitoring system. It performs merchant category analysis (left column) and lifestyle activity monitoring (right column) in parallel, and notifies 140 welfare officer terminals based on priority rankings (step 350, red, yellow, green). [Figure 4] This is a conceptual diagram of third-party detection using remittance network analysis. It shows a configuration in which a graph analysis module 460 detects concentrated remittance patterns (many-to-one) from multiple recipients to a specific account. [Figure 5] This is a flowchart for identity verification using liveness detection. It shows a configuration that proceeds from a dynamic action request at random timings (step 500) to identity verification via 3D facial recognition (step 530), and then to terminal locking (step 550) if authentication fails. [Figure 6] This is a logic diagram for limiting payments for highly liquid products. It shows a configuration that goes from determining whether a product is highly liquid (step 620) to determining a monthly threshold (step 630, e.g., 30,000 yen per month), and then, if the threshold is exceeded, executing a case worker approval request (step 640, "Request Case Worker Approval" in the diagram). [Figure 7] This is an example of a dashboard screen for welfare commissioners and caseworkers. It displays recipients in their assigned district categorized as normal (green, e.g., 70 people, 90%), caution (yellow, e.g., 6 people, 8%), and emergency (red, e.g., 2 people, 2%). The configuration includes detailed information on emergency recipients (last activity date and time, power outage, failure to meet medical appointments, etc.), response buttons (telephone, visit, medical, police), recent response history, and an automatic generation and transmission function for monthly reports. [Figure 8] This is a screen transition diagram for the smartphone app for benefit recipients. It shows the screen transitions from the home screen to the deposit notification, income type selection (salary, gift, etc.), confirmation submission, and history display, demonstrating that the entire declaration can be completed in under 24 seconds. [Figure 9] This is a flowchart for the integrated analysis of medical data and payment data. It shows a configuration that integrates medical records from systems such as the National Health Insurance claims system with payment data, and calculates risk scores for four patterns: detection of prescription drug resale, detection of organized fraud, detection of discrepancies in transportation expenses, and detection of fraudulent use of medical expenses. [Figure 10]This is a safety confirmation process by an AI voice agent. It shows a configuration where the first stage (0 - 2 hours) is AI voice call (step 1010), the second stage (2 - 6 hours) is notification to the regional support center (step 1030) and caseworker alert (step 1035, "CW Alerts" in the figure), and the third stage (over 6 hours) is automatic reporting to the emergency reporting service (step 1050), and all logs are recorded in the evidence repository 160 in sequence. [Figure 11] This is a configuration diagram of a safety confirmation system by integrating smart meters. It shows a configuration where lifeline usage data obtained from the power smart meter 191, gas smart meter 192, and water smart meter 193, as well as settlement history and data of the CBDC wallet 170, are comprehensively analyzed to distinguish high priority (power zero for 48 hours, water zero for 72 hours, and no settlement activities) and medium priority (normal power, no settlement for 72 hours, abnormal pattern), and then distributed to the dashboard alert (terminal 140) or AI voice agent (reference numeral 180) according to the priority.

Embodiments for Implementing the Invention

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. <Overall System Configuration>

[0015] As shown in FIG. 1, the fraud prevention and receipt system 100 of the present invention is composed of an information terminal 110 used by the recipient (an application installed on a smartphone), a plurality of payment servers 120 (including domestic payment APIs, overseas payment APIs, credit and debit card APIs), a welfare management server 130 of an administrative agency, a neighborhood committee terminal 140, an AI server 150, and an evidence repository 160.

[0016] The AI server 150 includes a data collection means 151, a determination means 152, a data fixation means 153, and an evaluation means 154. The communication between the information terminal 110 and the AI server 150 is encrypted, and the AI server 150 and the welfare management server 130 are connected by a secure line. The civil servant terminal 140 receives an alert from the determination means 152 and displays a dashboard screen (Figure 7). <Income declaration process that eliminates the declarant's burden for recipients>

[0017] The income declaration process shown in Figure 2 replaces the submission of the conventional monthly handwritten income declaration form. Conventionally, recipients had to create a handwritten income declaration form regardless of whether they had income, attach a copy of their salary statement or passbook, and bring it to or mail it to the welfare office. In this system, this series of procedures is completed within 24 seconds by the following automated process.

[0018] When the data collection means 151 detects a deposit into e-money or two-dimensional code settlement (step 200, for example, detection of a deposit of ￥8,000), the determination means 152 compares it with the declared income recorded in the welfare management server 130 (step 210, within 3 seconds required). As a result of the comparison, if a deposit that does not match the declared income is detected, a push notification is sent to the recipient's information terminal 110 (step 220, within 8 seconds from detection).

[0019] When the recipient taps the notification to launch the app (step 230), the determination means 152 displays the amount of the deposit, the name of the remitter, and the date and time in the declaration form with automatic input (step 240, corresponding to the notification screen in Figure 8). The recipient selects the type of income (salary, gift, etc.) shown and completes the declaration by tapping the send button only once (step 250, 23 seconds from the start).

[0020] After the declaration is completed, the data fixing means 153 records the response content, response date and time, and related transaction information in the evidence storage 160 in an immutable format including a hash value (step 260, all processing completed in 24 seconds). If the recipient does not respond within 48 hours (step 270), the recipient is automatically flagged as a target for investigation (step 280). <Dual monitoring to reduce the burden of visits by community welfare officers>

[0021] The dual monitoring system shown in Figure 3 is configured to understand the living conditions of recipients without requiring welfare officers to physically visit all assigned households. In the merchant category analysis in the left column, merchant categories are extracted from payment transactions (step 300) (step 310, data collection means 151), and if spending at gambling establishments or high-risk stores is detected (step 320), a warning notification is sent to the recipient and welfare officer (step 325).

[0022] In the right-hand column's lifestyle activity monitoring, continuous monitoring of payment patterns (step 330) is performed. If payment activity is interrupted for more than 48 hours (step 340), it is compared with smart meter data for electricity and water (step 345). Based on the comparison results, a priority ranking (red, yellow, green, step 350) is calculated and notified to the dashboard of the welfare officer terminal 140 (step 355). This allows welfare officers to focus their attention on households that truly require attention, rather than visiting all assigned households equally. <Remittance Network Analysis>

[0023] As shown in Figure 4, the determination means 152 uses the graph analysis module 460 to identify the correlation between transfers made from multiple recipients (users A (410), B (411), C (412), D (413), E (414)) to a specific suspicious account (400). The graph analysis module 460 detects many-to-one transfer patterns (organized inflow patterns), confirms the identity of the receiving account, and calculates the degree of confidence in organized fraud (e.g., 98%). If the degree of confidence exceeds the threshold, it notifies the administrative agency's terminal and immediately blocks the transfer. <Identity verification through liveness detection>

[0024] As shown in Figure 5, the determination means 152 generates a random challenge (step 500) that requests a dynamic action at irregular intervals and presents the recipient with liveness instructions such as blinking or shaking their head (step 510). When the recipient performs the action (step 520), depth analysis and anti-spoofing processing are performed by three-dimensional facial recognition (step 530). If authentication is successful, access is granted and a success log is recorded (step 545). If authentication fails three times, the terminal is locked (step 550) and a message indicating that manual identity verification is required is recorded. <Restrictions on payment of highly liquid goods>

[0025] As shown in Figure 6, the determination means 152 determines whether the payment target falls under the category of highly liquid goods including home appliances, precious metals, game equipment, or gift cards (step 620), and if so, determines whether it exceeds the monthly threshold (e.g., 30,000 yen per month) (step 630). If it is below the threshold, it is approved with a record (step 635), and if it exceeds the threshold, it executes an approval request from the caseworker (step 640). The caseworker is presented with the basis for the determination via the dashboard (Figure 7) and can approve or reject it (step 650). <Dashboard for Community Welfare Officers and Case Workers>

[0026] The dashboard shown in Figure 7 is a management screen that allows welfare commissioners and caseworkers to view the status of recipients under their care. At the top of the screen, recipients in their assigned district (e.g., District 1-2-3) are categorized into three levels: normal (green, e.g., 70 people, 90%), caution (yellow, e.g., 6 people, 8%), and emergency (red, e.g., 2 people, 2%).

[0027] For individuals requiring immediate attention (e.g., Taro Tanaka, 85 years old), the last activity date and time (e.g., 5 days ago) and alert details (e.g., zero electricity usage for 48 hours, missed medical appointment) are displayed, and buttons for phone calls, visits, medical institution contact, and police contact can be operated with a single touch.

[0028] The recent response history section in the middle of the screen displays a chronological record of recent responses in the assigned area (e.g., "February 15th | Tanaka | Call | No answer", "February 14th | Sato | Visit scheduled for tomorrow"). This response history can be used as handover material when a caseworker is transferred, ensuring that the recipient's situation is continuously monitored even when the person in charge changes.

[0029] The bottom of the screen features an automatic monthly report generation function. By operating a button, a welfare officer or caseworker can automatically generate a record of the current month's activities in PDF format and send it to the welfare office. This automates the process of creating monthly reports, which was previously done manually. <User interface design for the benefit recipient app>

[0030] The smartphone app shown in Figure 8 is designed to allow recipients to complete their income declarations with minimal effort. The home screen displays the recipient's status (e.g., normal), and when a deposit is detected, the deposit amount and sender are automatically displayed on the notification screen. Recipients simply select the type of income (e.g., salary, gift) and tap the submit button to complete the declaration. The declaration history screen allows users to check past declarations and their approval status. The entire declaration process is designed to be completed within 24 seconds. <Integrated Analysis of Medical Data>

[0031] As shown in Figure 9, the data collection means 151 acquires medical data from the National Health Insurance claims system, the eligibility verification system, and the electronic prescription system, and acquires payment data from the payment API, the online marketplace API, and the transportation API. The determination means 152 integrates these and executes the following four detection patterns: (1) Prescription drug resale detection: Detection of electronic payment within 72 hours after medical consultation. (2) Organized fraud detection: Detection of multiple beneficiaries visiting the same medical institution on the same day and receiving similar amounts of payment from each beneficiary. (3) Transportation expense inconsistency detection: Inconsistency between taxi fare claims and transportation IC card ride records. (4) Medical expense fraud detection: Electronic money charge within 48 hours after medical expense payment. These detection results are added together as a risk score and notified to the caseworker's terminal 140 when a threshold is exceeded. <An artificial intelligence voice agent to replace physical visits by community welfare officers>

[0032] The artificial intelligence voice agent shown in Figure 10 functions as a means of maintaining contact with recipients without requiring welfare officers to make direct visits. When a safety alert (step 1000) is issued for a recipient whose transaction history has been interrupted, the following steps are automatically executed.

[0033] In the first phase (0-2 hours), an AI voice agent automatically makes a call to the recipient (step 1010). If there is a response, a safety confirmation log is recorded (step 1025). If there is no response, the agent will recall up to four times (step 1020), and if there is no response for four consecutive times, it will proceed to the second phase.

[0034] In the second stage (2-6 hours), automatic notifications are sent to the local support center (step 1030) and the caseworker (step 1035, indicated as "CW Alerts" in the diagram). If the caseworker or local support center is able to confirm the recipient's safety, a log is recorded; if they are unable to confirm, the process moves to the third stage.

[0035] In the third stage (over 6 hours), a risk assessment (step 1040) is performed, and if a high risk is detected, an automatic notification to the emergency call service (step 1050) is initiated. Throughout all stages, all logs are recorded in evidence storage 160. This phased response significantly reduces the burden of physical visits by welfare commissioners, who are aging and facing a shortage of personnel, while still allowing for confirmation of the recipient's well-being. <Safety confirmation through smart meter integration>

[0036] As shown in Figure 11, the data collection means 151 obtains utility usage data at the recipient's place of residence via API from the electricity smart meter 191, the gas smart meter 192, and the water smart meter 193. The AI ​​server 150's composite analysis engine comprehensively analyzes this utility data along with payment history (payment API 120) and data from the CBDC wallet 170.

[0037] Two anomaly detection patterns are set: high priority (zero power usage for 48 hours, zero water usage for 72 hours, and no payment activity) and medium priority (normal power usage, no payment activity for 72 hours, and an anomaly pattern detected). In the case of high priority, the AI ​​voice agent 180 is activated (linked to the flow in Figure 10), and in the case of medium priority, an alert is displayed on the dashboard of the welfare officer terminal 140. In the case of low risk, logs are recorded in the evidence storage as part of continuous monitoring. <Relaxation of monitoring based on confidence scores>

[0038] If a recipient consistently makes accurate declarations for a predetermined period, the determination means 152 improves the recipient's confidence score. If the confidence score exceeds a predetermined threshold, the threshold for automatic approval in subsequent declarations is relaxed, and the automatic declaration process is executed according to predetermined classification criteria without requiring a response from the recipient. This further reduces the declaration burden on recipients who are using the system properly. <Data Minimization>

[0039] The information collected by data collection method 151 is limited to the amount, date and time, name of the remitter, merchant classification, and location information of payments necessary for proper benefit management. Details of purchased items or personal communications are excluded from collection. This ensures that only the information necessary to prevent fraudulent claims is collected while protecting the privacy of benefit recipients. <Detecting fraudulent patterns using machine learning>

[0040] The determination means 152 uses a machine learning model learned from past cases of fraudulent receipt to detect fraud prevention behavior patterns, including small, dispersed transfers, concentrated settlements during specific time periods, or repeated transfers to the same recipient. <Verification of living conditions using location information>

[0041] The data collection means 151 acquires location information at the time of payment, and the determination means 152 compares the location information with the place of residence or place of employment declared by the recipient. If the discrepancy between the two exceeds a predetermined standard, a fraud risk score indicating a contradiction with the actual living situation is calculated. [Explanation of Symbols]

[0042] 100 System to prevent fraudulent receipt of benefits 110 Information terminal (recipient's smartphone app) 120 Payment Servers (Payment API Group) 130 Welfare Management Server 140 Community Welfare Officer Terminal (Dashboard) 150 AI servers 151 Data Acquisition Methods 152 Judgment means 153 Data Fixing Means 154 Evaluation methods 160 Evidence Vault (WORM-type database, AES-256 encryption, hash chain verification) 170 CBDC Wallet System 180 AI Voice Agents 190 Smart Meter Management System 191 Electricity smart meter 192 Gas smart meter 193 Water smart meter

Claims

1. A system for preventing fraudulent receipt of benefits, which is connected via a network to an information terminal used by recipients of public assistance, payment servers managed by multiple payment service providers, and welfare management servers managed by administrative agencies, A data collection means that acquires transaction history across multiple payment services, including the recipient's bank account, electronic money, and QR code payment, and integrates the acquired transaction history in chronological order. A determination means that analyzes the transaction history acquired by the data collection means and, when it detects a deposit that has not been reported to an administrative agency, sends a notification to the information terminal requesting a response regarding whether or not the deposit is income eligible for calculation of the benefit amount. A data fixing means that receives the recipient's response to the aforementioned notification and records the response content, response date and time, and related transaction information in an immutable format including a hash value, A system for preventing fraudulent receipt of benefits, characterized by having the following features.

2. The system according to claim 1, characterized in that the data collection means acquires deposit history, charge history, person-to-person transfer history, and payment history for multiple electronic money and two-dimensional code payment services via an account aggregation platform provided by a financial services intermediary or an electronic payment service provider.

3. The system according to claim 1, characterized in that when the determination means detects a deposit to electronic money or a two-dimensional code payment, it automatically enters the amount of the deposit, the name of the sender, and the date and time into the declaration form and presents it to the information terminal, enabling the recipient to complete the declaration in a single operation.

4. The system according to claim 1, characterized in that the data fixing means records the recipient's response content together with a timestamp in a distributed ledger and stores it in a format that can be output as evidence at the time of administrative action.

5. The system according to claim 1, characterized in that the data collection means acquires location information at the time of payment, and the determination means compares the location information with the place of residence or place of employment declared by the recipient, and calculates a fraud risk score indicating a discrepancy with the actual living situation if the discrepancy between the two exceeds a predetermined standard.

6. The system according to claim 1, characterized in that the determination means notifies the terminal of a welfare officer or local monitoring officer that the recipient needs to be checked for their well-being when the transaction history is interrupted for a predetermined period of time or when a payment pattern that deviates significantly from the normal lifestyle pattern is detected.

7. The system according to claim 1, characterized in that the data collection means monitors deposits into bank accounts, charges to electronic money, withdrawals from ATMs, and deposits and withdrawals via person-to-person transfers, and the determination means subject all of these to determination of inconsistencies in income and expenditure.

8. The system according to claim 1, wherein the determination means detects fraud prevention behavior patterns, including small, dispersed transfers, concentrated settlements during specific time periods, or repeated transfers to the same recipient, using a machine learning model learned from past fraudulent benefit receipt cases.

9. The system according to claim 1, wherein the determination means identifies the correlation between transfers made from multiple recipients to a specific personal account by graph analysis, and notifies the administrative agency's terminal if there is suspicion of fraudulent misappropriation of benefits by a third party.

10. The system according to claim 1, characterized in that the determination means verifies the identity of the payment executor and the recipient using the biometric authentication function of the information terminal when a payment is executed or when an undeclared payment is detected, and if authentication is not achieved, records the transaction as a potentially fraudulent transaction and notifies the administrative agency.

11. The system according to claim 10, characterized in that the determination means performs liveness detection that requests dynamic action at irregular intervals, performs identity verification by facial recognition, fingerprint recognition, or voice recognition, and detects terminal occupancy by a third party.

12. The system according to claim 1, characterized in that the determination means sets an upper limit on the payment amount or requests approval from a welfare officer when the payment target is a highly liquid product including home appliances, precious metals, game equipment, or gift cards.

13. The system according to claim 1, characterized in that if the recipient continues to make proper declarations for a predetermined period, the system improves the recipient's confidence score and relaxes the threshold for automatic approval in subsequent declarations.

14. The system according to claim 1, characterized in that the information collected by the data collection means is limited to the amount of payment, date and time, name of the remitter, merchant classification, and location information necessary for proper benefit management, and details of purchased goods or personal communication content are excluded from the collection.

15. The system according to claim 1, wherein the data fixing means encrypts the collected data, allows access only to authorized persons, and records all access history.

16. The system according to claim 1, characterized in that the determination means automatically approves a predetermined percentage of all declarations based on the amount, frequency, confidence score, and presence or absence of evidence, thereby reducing the verification work of welfare officers.

17. The system according to claim 1, wherein the data collection means further acquires the medical institution's consultation history and prescription drug history, and the determination means notifies the terminal of the welfare officer if a payment via electronic payment or online marketplace is detected within a predetermined period from the consultation date, indicating that the transaction may be a resale of prescription drugs.

18. The system according to claim 17, characterized in that the determination means notifies the terminal of a welfare officer of a group of transactions that may be organized fraud in collusion with a medical institution when it is detected that multiple recipients visit the same medical institution on the same day and that the same or similar amounts of money are deposited into each recipient's electronic payment account within a predetermined period from the date of the visit.

19. The system according to claim 1, characterized in that the data collection means acquires the deposit and withdrawal history of a digital currency issued by a central bank through an interface provided by the management system of the digital currency, and the determination means includes the deposit and withdrawal of the digital currency as the subject of the balance sheet discrepancy determination.

20. The system according to claim 6, characterized in that, when a recipient is determined to require a safety check, an artificial intelligence voice agent automatically makes a phone call, analyzes the recipient's response using voice recognition, and if there is no response or an abnormality is detected, automatically sends a notification to a pre-set emergency contact.

21. The system according to claim 20, characterized in that it automatically executes a phased response based on the results of the safety confirmation, and sequentially executes the following at predetermined time intervals: firstly, a redial by an artificial intelligence voice agent; secondly, an automatic notification to a local support center; and thirdly, an automatic notification to an emergency call service.

22. The system according to claim 6, characterized in that the data collection means acquires lifeline usage data at the recipient's place of residence via an interface of a smart meter provided by a power company, gas company, or water utility, and the determination means performs a safety check by comprehensively analyzing the usage data and payment history.

23. The system according to claim 13, characterized in that, if the recipient's confidence score exceeds a predetermined threshold, the determination means automatically classifies the detected deposit according to a predetermined classification criterion, performs an automatic reporting process as income or non-income without requesting a response from the recipient, and subsequently notifies the recipient of the processing result to the recipient's information terminal.

24. The system according to claim 1, characterized in that the data collection means acquires peer-to-peer remittance history, digital item buying and selling history, virtual currency exchange history, and creator revenue receipt history within a social media platform, online game, virtual space service, or digital content distribution platform via an interface provided by the said platform, and the determination means includes these as the subject of income and expenditure discrepancy determination.

25. The system according to claim 19, wherein the data collection means obtains the deposit and withdrawal history, the exchange history between cryptocurrencies, and the deposit and withdrawal history of stablecoins in the recipient's cryptocurrency wallet via an interface provided by a cryptocurrency exchange, a decentralized exchange, or a wallet service, and the determination means includes the conversion to fiat currency, the receipt of stablecoins, or the profit from the sale of non-fungible tokens as the subject of the balance sheet inconsistency determination.

26. The system according to claim 10, characterized in that the determination means continuously analyzes the operation pattern, input rhythm, swipe motion characteristics, walking pattern, or daily location movement pattern of the information terminal using a machine learning model, and when a statistically significant deviation from the normal behavior pattern is detected, it requests additional user authentication as a possibility that the terminal may be used by someone other than the user.

27. The system according to claim 14, characterized in that the data collection means performs identity verification and eligibility verification of the recipient based on the technical standards for decentralized identifiers and verifiable credentials, and links data obtained from multiple payment services based on the recipient's consent using the decentralized identifier without relying on a centralized identification number system.

28. A method for processing information to prevent fraudulent receipt of public assistance benefits, The computer acquires transaction histories across multiple payment services, including those from the recipient's bank account, electronic money, and QR code payments, and integrates the acquired transaction histories in chronological order. The computer analyzes the integrated transaction history and, upon detecting a deposit that has not been reported to the administrative agency, sends a notification to the recipient's information terminal requesting a response regarding whether or not the deposit is income eligible for calculating the benefit amount. The computer receives the recipient's response to the notification and records the response content, response date and time, and related transaction information in an immutable format including a hash value. An information processing method characterized by including

29. The method according to claim 28, characterized in that, in the integration step, the deposit and withdrawal history of a digital currency issued by a central bank is further obtained through an interface provided by the management system of the digital currency, and in the analysis step, the deposit and withdrawal of the digital currency is included in the determination of the balance of accounts.

30. The method according to claim 28, further comprising the steps of: after the recording step, the computer automatically making a phone call to a recipient whose transaction history has been interrupted for a predetermined period using an artificial intelligence voice agent, analyzing the recipient's response using voice recognition, and if there is no response or an abnormality is detected, sequentially performing a step-by-step response including making a redial, notifying a local support center, and notifying an emergency call service.

31. The method according to claim 28, characterized in that, in the integration step, further lifeline usage data at the recipient's place of residence is obtained through the interface of a smart meter provided by a power company, gas company, or water utility, and in the analysis step, the usage data and payment history are combined and analyzed to perform a safety check.

32. The method according to claim 28, characterized in that, in the analysis step, if the recipient's confidence score exceeds a predetermined threshold, the detected deposit is automatically classified according to a predetermined classification criterion, an automatic reporting process is performed as income or non-income without requesting a response from the recipient, and the processing result is subsequently notified to the recipient's information terminal.

33. A computer-readable non-temporary recording medium having recorded a program for causing a computer to perform the method described in any one of claims 28 to 32.

Images