Network Centrality Metrics for Key Entity Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying key opinion leaders in fields like medicine are inefficient and biased, relying on time-consuming and costly surveys, which do not objectively select influential individuals.
Innovation Solution
A method using network centrality metrics and statistical computer programs to identify and segment key entities in a network, assigning rank-based scores and calculating reach to form a subgraph of key entities, providing an alternative to survey-based approaches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If surveys are used to identify key opinion leaders, then individuals can be selected based on their opinions, but the process becomes time-consuming and biased
Solution Approach 1:
The patent replaces the mechanical survey process with automated network analysis algorithms. Instead of manually collecting and analyzing survey responses, the system uses computer programs to automatically calculate network centrality metrics (degree, betweenness, eigenvector centrality) for all entities in a network, thereby eliminating the time-consuming and biased survey methodology while maintaining objective identification of key entities
Solution Approach 2:
The network analysis system performs self-service by automatically processing network data and generating key entity lists without requiring external survey respondents. The system uses existing network relationship data and computational algorithms to independently identify key entities, eliminating the need for human participants to self-report their opinions
2Measurement precision
If surveys are used to identify key opinion leaders, then selections can be made, but the process becomes expensive and requires monetary compensation
Solution Approach 1:
The patent replaces the costly survey system with free automated network analysis. Instead of paying respondents for their time and opinions, the system uses computational algorithms to process existing network data and automatically generate objective rankings of key entities, thereby eliminating monetary costs while maintaining measurement precision
Solution Approach 2:
The system uses disposable computational resources and algorithms rather than expensive survey infrastructure. The network analysis can be performed using standard computer programs and existing data, eliminating the need for expensive survey administration, printing, distribution, and compensation systems
3Measurement precision
If network centrality metrics are calculated for all entities, then comprehensive key entity identification is achieved, but computational complexity increases
Solution Approach 1:
The patent segments the network analysis process into distinct computational steps: calculating individual centrality metrics (degree, betweenness, eigenvector centrality) for each entity, ranking entities based on these metrics, and filtering to identify key entities. This segmentation allows comprehensive analysis of all entities while managing computational complexity through systematic processing
Solution Approach 2:
The system manages computational complexity by changing parameters such as the number of centrality metrics calculated, the threshold values for identifying key entities, and the size of the network being analyzed. By adjusting these parameters, the system can comprehensively analyze large networks while controlling computational resource requirements
Data Source
AI summary
A method, system and computer-program product for generating a subgraph of key entities in a network and organizing entities in the subgraph are disclosed. The technique uses social network analysis centrality metrics to identify key entities in a network. The technique also uses social network analysis centrality metrics to categorize key entities into different types.