Delayed payment settlement anti-fraud system based on suspected fraud-related account

Abstract

The invention discloses a delayed payment settlement anti-fraud system based on a suspected fraud-related account. The anti-fraud system comprises an API layer, a service layer and a data layer, the API layer comprises an anti-fraud real-time interception API and unified standard anti-fraud data source API integration; the service layer comprises a risk delay control center, a feedback decision and an anti-fraud data center; and the data layer comprises an anti-fraud data center and an anti-fraud data graph. Through real-time data stream processing, a rule engine and a delayed settlement mechanism, delayed settlement risk management and control are carried out on transactions, and an automatic and manual compensation feedback mechanism is adopted to guarantee user transaction experience to the greatest extent.

Description

Technical Field

[0001] The present invention relates to the field of financial anti-fraud, and in particular to an anti-fraud system for delayed payment settlement based on suspected fraud-related accounts. Background Art

[0002] With the rapid development of financial technology, electronic payment has become more and more popular. The detection and management of suspected fraudulent accounts is an important part of the modern financial anti-fraud system. The delayed payment settlement anti-fraud system introduces a buffer time to prevent possible illegal transfers and payments when suspected fraudulent behavior is detected, and protects the user transaction experience through automatic and manual compensation feedback mechanisms, thereby greatly reducing financial risks.

[0003] The existing solutions that use various machine learning algorithm models to identify suspected fraudulent account transactions and then block them have the following shortcomings. First, risk identification through models alone is relatively simple, and the models are prone to false positives and false negatives. On the other hand, the optimization cost of the models is relatively high, and the real-time performance is not good. Direct blocking will damage the user's transaction experience, prevent transactions from being carried out in a timely manner, and harm the interests of legitimate users. Summary of the invention

[0004] In view of the above problems, the present invention is proposed to provide an anti-fraud system for delayed payment settlement based on suspected fraudulent accounts, which overcomes the above problems or at least partially solves the above problems.

[0005] According to one aspect of the present invention, there is provided an anti-fraud system for delayed payment settlement based on suspected fraudulent accounts, the anti-fraud system comprising: an API layer, a service layer and a data layer;

[0006] The API layer includes anti-fraud real-time interception API and unified standard anti-fraud data source API integration;

[0007] The service layer includes risk delay management, feedback decision and anti-fraud data center;

[0008] The data layer includes an anti-fraud data center and an anti-fraud data map.

[0009] Optionally, the process of the anti-fraud data center includes:

[0010] Anti-fraud data release;

[0011] Data cleaning and confidence setting;

[0012] Persist data in RDB in multiple data dimensions;

[0013] Perform model training and inference.

[0014] Optionally, the data dimensions include: account, device and IP.

[0015] Optionally, the risk delay control specifically includes:

[0016] User initiates a transaction;

[0017] Obtain anti-fraud data from the data center through multiple dimensions of information on both parties to the transaction;

[0018] Based on the anti-fraud data, quickly identify whether the transaction involves fraud through rule engines, credibility and other mechanisms;

[0019] Data flows back into the algorithm model;

[0020] If fraud is involved, the delayed payment settlement mechanism will be triggered;

[0021] Set a delay time window based on credibility and trigger the feedback process;

[0022] Record the anti-fraud subject information of the data source.

[0023] Optionally, the feedback decision process includes:

[0024] Implement notification to users;

[0025] User feedback;

[0026] The system automatically generates a user's transaction data map;

[0027] Further decision-making and processing are carried out through the expert system to obtain processing results;

[0028] The processing result is sent to the user.

[0029] The present invention provides an anti-fraud system for delayed payment settlement based on suspected fraud accounts, the anti-fraud system includes: an API layer, a service layer and a data layer; the API layer includes an anti-fraud real-time interception API and a unified standard anti-fraud data source API integration; the service layer includes risk delay management, feedback decision and anti-fraud data center; the data layer includes an anti-fraud data center and an anti-fraud data map. Through real-time data stream processing, rule engine, and delayed settlement mechanism, the transaction is delayed and settled risk management is carried out, and automatic and manual compensation feedback mechanisms are used to maximize the user transaction experience.

[0030] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0032] Figure 1 A block diagram of a delayed payment settlement anti-fraud system based on suspected fraudulent accounts provided by an embodiment of the present invention;

[0033] Figure 2 A method flow chart corresponding to a delayed payment settlement anti-fraud system based on a suspected fraud-related account provided by an embodiment of the present invention;

[0034] Figure 3 A timing diagram of an anti-fraud data center provided by an embodiment of the present invention;

[0035] Figure 4 A timing diagram of a delay control mechanism provided by an embodiment of the present invention;

[0036] Figure 5 A timing diagram of the feedback decision capability provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0038] The terms "comprises" and "having" and any variations thereof in the description embodiments, claims and drawings of the present invention are intended to cover non-exclusive inclusions, for example, including a series of steps or units.

[0039] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0040] like Figure 1 As shown in the figure, the core of the system has three layers, namely API layer, service layer, and data layer. The API layer includes anti-fraud real-time interception API and unified standard anti-fraud data source API integration. The service layer mainly includes risk delay control, feedback decision, and anti-fraud data center. The data layer mainly includes anti-fraud data center (RDB) and anti-fraud data graph (GraphDB).

[0041] like Figure 2As shown, the present invention achieves technical goals through three core capabilities, namely: anti-fraud data center, delay control mechanism, and feedback decision-making capability.

[0042] 1. Anti-fraud data center

[0043] On the one hand, it uses machine learning algorithm models for offline production, and on the other hand, it uses the unified standard anti-fraud data source API integration capability to obtain highly reliable data from multiple anti-fraud data source platforms. The final data is maintained in a unified anti-fraud data model to facilitate subsequent real-time risk management.

[0044] 2. Delay control mechanism

[0045] For real-time transactions of users, through the rule engine, credibility and other mechanisms, it is quickly identified whether the transaction involves fraud. If it involves fraud, the delayed payment settlement mechanism is triggered, the delay time window is set through credibility, and the feedback process is triggered. At the same time, the anti-fraud subject information of the data source is recorded. In addition, the data will also flow back to the model.

[0046] 3. Feedback decision-making ability

[0047] The system automatically generates a user's transaction data map and further makes decisions through the expert system. If it is invalid, the transaction will be processed normally, otherwise the transaction will be blocked. On the other hand, users will also receive real-time notifications and actively provide feedback.

[0048] The timing diagram of the anti-fraud data center is as follows Figure 3 As shown:

[0049] Regarding high-trust data processing, the system receives anti-fraud data released by external data sources, performs data cleansing, sets confidence levels, and persists the data in RDB in multiple dimensions, including accounts, devices, IP addresses, etc. Relational storage facilitates real-time management and control of subsequent risks.

[0050] Regarding model data integration, the core is to sample training data from the selected transaction data, train it through various models, finally select the model with the best effect, and then evaluate and infer the actual data, and persist the output results in RDB.

[0051] The timing diagram of the delay control mechanism is as follows Figure 4 As shown:

[0052] When a user initiates a transaction, anti-fraud data is obtained from the data center through multi-dimensional information of both parties to the transaction. Based on the data from the previous step, the rule engine, credibility and other mechanisms are used to quickly identify whether the transaction involves fraud. At the same time, the data is fed back to the algorithm model. If it involves fraud, the delayed payment settlement mechanism is triggered, the delay time window is set through credibility, and the feedback process is triggered to record the anti-fraud subject information of the data source.

[0053] Feedback decision-making ability

[0054] The system automatically generates a user's transaction data map and further makes decisions through the expert system. If it is invalid, the transaction will be processed normally, otherwise the transaction will be blocked. On the other hand, the user will also receive real-time notifications and actively provide feedback. The timing diagram of the feedback decision-making ability is as follows Figure 5 As shown:

[0055] Beneficial effects: The present invention uses rich API integration capabilities to supplement high-quality suspected fraud data sets on top of the model, builds a rich high-quality anti-fraud data center, and provides a solid data chassis for transaction security. On the other hand, through real-time data stream processing, rule engine, and delayed settlement mechanism, the transaction is delayed and settled risk control is carried out, and automatic and manual compensation feedback mechanisms are used to maximize the user's transaction experience.

[0056] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A delayed payment settlement anti-fraud system based on suspected fraud accounts, characterized in that: The anti-fraud system includes: an API layer, a service layer, and a data layer; The API layer includes anti-fraud real-time interception API and unified standard anti-fraud data source API integration; The service layer includes risk delay management, feedback decision and anti-fraud data center; The data layer includes an anti-fraud data center and an anti-fraud data map.

2. According to claim 1, a delayed payment settlement anti-fraud system based on suspected fraud accounts is characterized in that: The process of the anti-fraud data center includes: Anti-fraud data release; Data cleaning and confidence setting; Persist data in RDB in multiple data dimensions; Perform model training and inference.

3. According to claim 2, a delayed payment settlement anti-fraud system based on suspected fraud accounts is characterized in that: The data dimensions include: account, device and IP.

4. According to claim 1, a delayed payment settlement anti-fraud system based on suspected fraud accounts is characterized in that: The risk delay control specifically includes: User initiates a transaction; Obtain anti-fraud data from the data center through multiple dimensions of information on both parties to the transaction; Based on the anti-fraud data, quickly identify whether the transaction involves fraud through rule engines, credibility and other mechanisms; Data flows back into the algorithm model; If fraud is involved, the delayed payment settlement mechanism will be triggered; Set a delay time window based on credibility and trigger the feedback process; Record the anti-fraud subject information of the data source.

5. According to claim 1, a delayed payment settlement anti-fraud system based on suspected fraud accounts is characterized in that: The feedback decision process includes: Implement notification to users; User feedback; The system automatically generates a user's transaction data map; Further decision-making and processing are carried out through the expert system to obtain processing results; The processing result is sent to the user.

Images