Cross-Service Fraud Estimation via User Data Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fraud estimation systems are limited in precision as they only detect users whose activities resemble those on a blacklist, failing to effectively identify fraudulent users with different patterns of behavior.
Innovation Solution
A fraud estimation system that compares user information across multiple services, utilizing relevance information to estimate fraudulence based on similarities and differences in user data, thereby enhancing detection precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fraud detection is based on comparing action histories with blacklist users, then users with similar fraudulent patterns can be detected, but fraud estimation precision cannot be raised to a satisfactorily high level because fraudulent users have diverse behaviors that do not always resemble blacklist patterns
Solution Approach 1:
The patent segments the fraud detection process into multiple independent comparison dimensions: action history comparison, user information comparison, and service relevance comparison. Each dimension is handled separately by different obtaining means (action history obtaining means, user information obtaining means, relevance information obtaining means), allowing the system to evaluate diverse fraudulent patterns through multiple specialized channels rather than a single rigid comparison method
Solution Approach 2:
The patent adds a new dimension to fraud detection by introducing service relevance information comparison. Instead of only comparing action histories in one dimension, the system now compares user information across multiple services while considering the relevance relationships between different services. This multi-dimensional approach enables detection of fraudulent users who may have different action patterns but share common user information characteristics across services
2Measurement precision
If multiple services are integrated for fraud estimation, then detection precision can be improved, but system complexity increases due to managing multiple services and their relevance information
Solution Approach 1:
The patent creates a universal fraud estimation framework that can handle multiple services through a standardized interface. The service providing system acts as a universal platform that receives requests from various services, manages their relevance information, and returns fraud estimation results. This multi-functional design allows the system to integrate diverse services without proportionally increasing complexity, as the core fraud estimation logic remains consistent across different service types
Solution Approach 2:
The patent introduces an intermediary layer (the service providing system) that mediates between individual services and the fraud estimation logic. This intermediary manages the complexity of multiple services by standardizing data formats, handling relevance information relationships, and coordinating comparisons across services. The intermediary absorbs the system complexity, allowing individual services to remain simple while achieving integrated fraud detection precision
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Relevance information obtaining means (104c) of a fraud estimation system (S) is configured to obtain relevance information about relevance between one service and another service. Comparison result obtaining means (105c) is configured to obtain a comparison result of a comparison between user information of a target user in the one service and user information of a fraudulent user or an authentic user in the another service. Estimation means (106c) is configured to estimate fraudulence of the target user based on the relevance information and the comparison result.