Trust Score and Deviation for New User Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for detecting abusive users in transactional systems are ineffective for new users without a history of interactions, often resulting in false positives and compromising both the operator and legitimate users.
Innovation Solution
A decision-making assist method that determines a trust score and trust deviation for users based on their integration into user communities, using community detection algorithms and social proximity analysis to assess the reliability of interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If black lists or white lists are used to detect abusive users, then the detection of clearly identified fraud sources is improved, but the effectiveness against new users without interaction history deteriorates
Solution Approach 1:
The system pre-computes and stores trust scores and social proximity metrics for users based on their interaction history before they become targets of communication requests. This preliminary analysis enables the system to quickly assess new users without requiring real-time computation when a communication request is received, thereby improving both detection accuracy and adaptability to new users.
Solution Approach 2:
The patent transitions from traditional one-dimensional black/white list approaches to a multi-dimensional assessment system that considers social proximity, trust scores, and community relationships. By adding these additional dimensions of analysis, the system can effectively evaluate users regardless of whether they have extensive or no interaction history, resolving the contradiction between detection accuracy and adaptability to new users.
2Stability of the object's composition
If social distance computing is applied to new users, then the theoretical framework is maintained, but false positives increase because social profile is similar to abusive users
Solution Approach 1:
The system pre-computes trust scores and social proximity metrics for all users during their interaction history, creating a buffer of pre-analyzed data. When a new user receives a communication request, the system uses these pre-computed values rather than computing social distance in real-time, thereby cushioning against the false positives that would otherwise occur due to the limitations of real-time social distance computation for new users.
Solution Approach 2:
The patent introduces trust scores and social proximity metrics as intermediary indicators between the raw interaction data and the final abuse detection decision. These intermediaries provide a more reliable basis for assessment than direct social distance computation, especially for new users, thereby maintaining framework consistency while improving detection reliability and reducing false positives.
3Reliability
If comprehensive user monitoring is implemented to improve security, then the detection of abusive users is improved, but the complexity of the system increases
Solution Approach 1:
The system automatically computes and updates trust scores and social proximity metrics based on user interaction data without requiring manual intervention or complex configuration. The algorithms self-adjust as new interactions occur, maintaining high security levels while minimizing the operational complexity and administrative burden on the system.
Data Source
AI summary
A decision-making assist method for a user of a first terminal receiving a communication request originating from a second terminal, the user of the second terminal being unknown to the user of the first terminal. The method includes determining a trust score of the user of the second terminal and a trust deviation, the trust deviation providing information on a reliability of the trust score. The trust score and the trust deviation are determined from a list including at least one user of a third terminal to which the second terminal has already transmitted a communication request, the at least one user of the third terminal belonging to at least one user community.