Bias Detection in Online Reviews via Relationship Strength Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for detecting biased online reviews rely on self-regulation within user communities, failing to account for external information such as personal or professional ties between reviewers and entities under review, which can lead to undetected bias.
Innovation Solution
A computer system that estimates relationship strengths between reviewers and entities by analyzing co-occurrence in Web documents, using search queries to calculate confidence levels of association rules, and identifying both direct and indirect associations to detect potential bias.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If self-regulation within user communities is used to detect biased reviews, then the system is simple to operate and requires minimal external resources, but it fails to detect bias arising from external relationships between reviewers and entities
Solution Approach 1:
The patent introduces an intermediary component that acts as a bridge between the user community self-regulation system and external information sources. This intermediary systematically collects and processes external data about reviewer-entity relationships, then feeds this information back to the review monitoring system. By doing so, it enables the detection of biased reviews based on external relationships without requiring complete system redesign, thus improving bias detection accuracy while managing complexity through a modular approach.
2Measurement precision
If external information about reviewer relationships is incorporated into bias detection, then detection accuracy improves, but information processing complexity and computational resources increase
Solution Approach 1:
The patent extracts only the specific external information needed for bias detection - namely, relationships between reviewers and entities under review - rather than processing all available external data. By selectively extracting relevant relationship data and filtering out unnecessary information, the system achieves improved bias detection accuracy while minimizing information processing complexity and computational resource requirements.
3Measurement precision
If comprehensive search queries are issued to estimate relationship strength, then relationship detection accuracy improves, but time consumption and computational resources increase
Solution Approach 1:
The patent applies partial action by issuing search queries that are sufficient to achieve acceptable relationship strength estimation accuracy without being overly comprehensive. Rather than exhaustively searching all possible information sources with maximum detail, the system performs targeted searches that capture the essential relationships, thereby reducing processing time and computational resource consumption while maintaining adequate detection accuracy.
Data Source
AI summary
One embodiment provides a computer system for detecting associations between a reviewer and an entity under review. During operation, the system estimates a relationship strength between the reviewer and the entity under review, and determines whether the relationship strength between the reviewer and the entity under review exceeds a predetermined threshold.