Batch Transaction Risk Control via Graph Indicators
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current risk control systems for electronic payments are ineffective in identifying risks in distributed batch transactions, leading to potential financial losses as they are designed to assess single transactions rather than batch transactions, missing collective risks and failing to prevent malicious activities.
Innovation Solution
A method and device that categorize batch transaction data based on attributes, generate graph indicators to identify risks, and input these indicators into risk identification models to determine the presence of risks, enabling quick and accurate risk identification in batch transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a risk control system conducts risk prevention and control for single electronic payments, then real-time risk determination and decision-making can be performed, but distributed batch electronic payments cannot be identified and risks are missed
Solution Approach 1:
The patent segments batch transaction data into multiple categories based on first attributes (such as transaction types, user groups, or time periods). Each category is then processed separately through graph construction and risk identification, allowing the system to handle different transaction patterns with appropriate analysis methods while maintaining overall batch transaction coverage.
Solution Approach 2:
The patent transitions from analyzing single transactions in isolation to constructing graphs that represent multi-dimensional relationships among transactions. By adding dimensional information about transaction networks, temporal patterns, and categorical groupings, the system can identify batch risks that single-transaction analysis would miss, thereby improving both accuracy and batch transaction coverage.
2Measurement precision
If graph indicators are generated for each category of batch transaction data, then collective risks can be identified accurately, but system complexity increases
Solution Approach 1:
The patent divides batch transaction data into multiple categories based on first attributes, and generates graph indicators separately for each category. This segmentation allows complex batch transaction analysis to be broken down into manageable categorical subsets, improving collective risk identification accuracy while organizing system complexity through structured categorization.
Solution Approach 2:
The patent employs a universal graph construction framework that can process different categories of transaction data through the same core algorithmic steps. The graph indicator generation mechanism serves multiple functions across different transaction categories, reducing the need for separate complex systems for each transaction type while maintaining accurate collective risk identification.
Data Source
AI summary
A method for data risk control comprises categorizing batch transaction data in a preset time period according to a first attribute, generating a graph indicator of a corresponding graph for each category of the batch transaction data according to a second attribute, the corresponding graph configured to identify risks in the batch transaction data, inputting the graph indicators corresponding to different categories of the batch transaction data into corresponding risk identification models, and determining whether the batch transaction data corresponding to the input graph indicators has a risk based on results output by the models. This scheme can quickly and accurately identify risks in batch transaction data.