Adaptive Network Graph Display for Contact Tracing Links
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video analytics systems struggle to efficiently identify highly related potential associates from a complex network, making it difficult to trace disease transmission links during pandemics, especially when dealing with large networks of co-appearances.
Innovation Solution
A method and apparatus for adaptively displaying potential subjects and target subjects by receiving an input, identifying potential subjects based on a threshold of relatedness, and displaying them with edges indicating relationship levels, using a contact tracing server and remote assistance servers to manage and visualize the network graph.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a complete network graph of all subjects is displayed, then all potential associates are visible, but the complexity of the network makes it difficult to identify highly related potential associates
Solution Approach 1:
The network graph is segmented into multiple levels based on relationship strength. The system divides all subjects into different categories (e.g., highly related, moderately related, weakly related) and displays them in hierarchical layers, allowing users to focus on the most relevant connections first while maintaining the option to explore less significant relationships.
Solution Approach 2:
Different visual properties are applied to different parts of the network graph based on their relationship strength to the target subject. Highly related subjects are displayed with prominent visual characteristics (e.g., larger nodes, brighter colors), while less related subjects have subdued visual properties, enabling users to quickly identify important connections without being overwhelmed by the entire network.
2Measurement precision
If manual analysis of large networks is performed, then comprehensive data review is possible, but the time and effort required to detect transmission links increases significantly
Solution Approach 1:
The system performs preliminary automated analysis of the network graph before user interaction, pre-calculating and organizing subjects by their relationship strength to the target. This preliminary processing creates a pre-structured visualization that immediately highlights potential transmission links, eliminating the need for users to manually explore the entire network and significantly reducing analysis time while maintaining detection accuracy.
Solution Approach 2:
The manual mechanical process of reviewing network connections is replaced with an automated computational system that algorithmsically analyzes relationships, calculates connection strengths, and generates the hierarchical visualization. This substitution of manual analysis with automated computing preserves the thoroughness of comprehensive review while dramatically reducing the time and human effort required.
3Loss of information
If all subjects in the network are displayed with equal visibility, then no information is hidden, but the specific potential associates highly related to the target subject become difficult to uncover
Solution Approach 1:
The display system is made dynamic and interactive, allowing users to adjust the level of detail and relationship threshold according to their needs. Users can dynamically filter the network to show only highly related subjects when needed, or expand the view to include less significant connections, maintaining both information completeness and identification precision through flexible, user-controlled visualization.
Data Source
AI summary
According to an a first aspect, there is a method of adaptively displaying at least one potential subject and a target subject, the method comprising receiving an input selecting the target subject from a network graph including a plurality of subjects who appear in the proximity as the target subject (204); identifying the at least one potential subject based on a threshold, the threshold indicating the minimum level of relatedness from the target subject (206); and displaying the at least one potential subject, the target subject and an edge between each of the at least one potential subject and the target subject, the edge indicating a level of relationship between the at least one potential subject and the target subject (208).