Visual Data Entity Relationship Detection via Multi-Dimensional Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods fail to effectively detect non-obvious relationships between entities from visual data sources, which is crucial for fraud detection and other applications, as they rely on digital and text data, making it difficult to uncover connections masked by techniques like layering and higher degrees of separation.
Innovation Solution
A processor calculates co-occurrence frequency and distance proximity scores, determines event types and timeline relationships from visual data, and combines these to establish a relationship score for detecting non-obvious connections between entities, using visual data sources such as images and videos.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods using digital and text data are used for relationship detection, then the detection process is simple, but the ability to uncover non-obvious relationships is insufficient
Solution Approach 1:
The patent transitions from traditional text-based relationship detection to multi-dimensional visual data analysis. By incorporating spatial coordinates, temporal information, and visual features from images and videos, the system detects relationships in additional dimensions beyond simple text co-occurrence, thereby uncovering non-obvious relationships that traditional methods miss.
Solution Approach 2:
The detection process is segmented into multiple independent analysis components: visual data processing extracts entities and their attributes, spatial analysis calculates distance metrics, temporal analysis determines timeline relationships, and event type classification identifies interaction contexts. Each segment processes specific aspects of visual data independently, then integrates results to achieve comprehensive relationship detection.
2Reliability
If visual data sources are analyzed to detect non-obvious relationships, then fraud detection capability is improved, but computational complexity increases
Solution Approach 1:
The visual data processing system performs multiple functions simultaneously: entity recognition, attribute extraction, spatial relationship calculation, temporal sequencing, and event classification. This multi-functional approach consolidates what would otherwise require separate processing systems, improving fraud detection reliability while managing computational complexity through integrated processing.
Solution Approach 2:
The system transforms visual data into standardized relational parameters including distance metrics, temporal intervals, and event type codes. By converting diverse visual information into uniform parameter formats, the system enables consistent relationship scoring and comparison across different data sources, enhancing detection reliability without proportionally increasing processing complexity.
3Measurement precision
If multiple factors (co-occurrence frequency, distance proximity, event type, timeline relationship) are considered in relationship scoring, then detection accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary processing of visual data to pre-extract entities, their attributes, and basic relationships before relationship scoring. By preparing and organizing visual data in advance with pre-identified entities and their properties, the system reduces the computational burden during the actual relationship analysis phase, maintaining high scoring accuracy while reducing overall processing time.
Data Source
AI summary
In an approach for detecting non-obvious relationships between entities from visual data sources, a processor calculates a co-occurrence frequency score for an entity pair from visual data. A processor calculates a distance proximity score for the entity pair from the visual data. A processor determines an event type in the visual data. A processor determines a timeline relationship in the visual data. A processor calculates a relationship score based on the co-occurrence frequency score, the distance proximity score, the event type, and the timeline relationship. A processor detects a relationship between the entity pair based on the relationship score.