Pair Risk Detection for Fraudulent User-Content Provider Links
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing click fraud detection methods fail to identify fraudulent user-content provider pairs across various attack techniques, including host-based, proxy-based, botnet-based, and coalition-based attacks, as they primarily focus on individual entities rather than the relationship between users and content providers.
Innovation Solution
A method and system for identifying fraudulent user-content provider pairs by determining initial risk values for both entities, generating functional representations based on relational data, and converging these values to detect suspicious activity, applying a fraudulent label to pairs exceeding a threshold risk value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing click fraud detection methods focus on individual entities (users or content providers separately), then the detection system is simpler to implement, but the ability to identify fraudulent user-content provider pairs across various attack techniques is insufficient
Solution Approach 1:
The patent merges separate detection approaches for users and content providers into a unified pair-based detection system. By simultaneously analyzing and updating risk values for both users and content providers based on their interactions, the system achieves more comprehensive fraud detection that captures collaborative attacks while maintaining manageable complexity through integrated processing
Solution Approach 2:
The system implements feedback mechanisms where the risk value of a user influences the risk value of associated content providers and vice versa. This iterative feedback process allows the system to progressively refine its detection accuracy by continuously updating risk assessments based on observed interactions, thereby improving reliability without requiring overly complex initial system design
2Measurement precision
If the system applies strict fraud detection thresholds to prevent false positives, then the precision of fraud identification improves, but the ability to detect subtle coordinated fraud patterns decreases
Solution Approach 1:
The patent employs dynamic risk value updates that adapt to different attack patterns. Rather than using fixed thresholds, the system continuously adjusts risk assessments based on the observed behavior patterns, allowing it to detect subtle coordinated fraud while maintaining precision. The risk values evolve dynamically as new interaction data becomes available, enabling the system to adapt to emerging attack techniques
Solution Approach 2:
The system changes detection parameters adaptively by updating risk values based on interaction history and patterns. This allows the detection mechanism to adjust its sensitivity and thresholds dynamically, improving both precision in identifying true fraud and adaptability to various attack patterns without requiring manual recalibration
Data Source
AI summary
The present teaching generally relates to identifying fraudulent content provider-user device pairs. In one embodiment, an initial user risk value and an initial content provider risk value may be determined. A first functional representation of a user risk value may be generated based on the initial user risk value and relational data. A second functional representation of a content provider risk value may be generated based on the initial content provider risk value and the relational data. A converged user risk value and a converged content provider risk value associated with the first and second representations converging may be determined. A pair risk value may be determined based on the converged user risk value and the converged content provider risk value. A fraudulent label may then be applied to interaction events detected by the content provider from the user in response to the risk pair value satisfying a condition.