Identity Information Identification via Dynamic Association Risk Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods are inadequate in identifying fraudulent use of identity information, especially when the usage environment changes, as they rely on static account behavior and fail to effectively detect fraudulent activity in dynamic scenarios.
Innovation Solution
A method and apparatus that analyze association relationships between identity information and account information, determining a risk value by establishing connections based on common features such as shared interactions, objects, or devices, allowing for the identification of potentially fraudulent identity information even when the usage environment changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static account behavior analysis is used to identify fraudulent identity information, then the identification method is simple, but it fails to detect fraudulent activity when usage environment changes
Solution Approach 1:
The patent transforms the static fraud detection approach into a dynamic one by continuously monitoring account behavior patterns and comparing them against learned normal patterns. The system adapts to changing usage environments by detecting deviations from established patterns rather than relying on fixed rules, enabling reliable fraud detection even when legitimate usage patterns evolve.
Solution Approach 2:
The system implements feedback mechanisms where account behavior data is continuously collected, analyzed, and used to refine fraud detection models. The comparison between actual behavior patterns and normal patterns creates a feedback loop that improves detection accuracy over time, allowing the system to learn from new fraud patterns while maintaining simplicity through automated pattern recognition.
2Reliability
If association relationships between multiple accounts are analyzed, then fraud detection capability is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent segments the complex analysis of account associations into manageable components by focusing on specific behavioral patterns and interaction types. Instead of analyzing all possible account relationships simultaneously, the system divides the problem into discrete pattern recognition tasks, processing only relevant associations that match known fraud indicators, thereby reducing overall processing time while maintaining detection capability.
Solution Approach 2:
The system applies partial action by selectively analyzing only the most critical association patterns rather than exhaustively examining all account relationships. By focusing computational resources on high-risk pattern types and using heuristics to prioritize analysis, the system achieves effective fraud detection with reduced processing overhead compared to comprehensive analysis of all account interactions.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A method and an apparatus for identifying identity information are provided in embodiments of the present invention, wherein the method comprises: acquiring user data, the user data comprising identity information and account information of a user; establishing an association relationship between account information and identity information that are bound in the user data, and establishing an association relationship between two pieces of account information having common features in the user data; and determining a risk value of target identity information in the user data according to the established association relationships, and determining, according to the determined risk value, whether the target identity information has a risk of being used fraudulently. The technical solution described in this specification can improve the efficiency of identifying identity information.