A Network Model for Ranking Groups by Their Actions

To understand interactions between entities (for example, people, objects or groups) systems of interactions are often modeled as graphs linking nodes (entities) with edges representing various types of connections between the entities. After data collection there are many statistical approaches to analyzing the data. Our approach is to model data as a graph and explore the graph using a variety of tools to reveal latent structure or patterns. Here, we propose a metric that ranks groups based on their actions toward or their interactions with others. It applies to a variety of applications, its mean and variance can be computed quickly, and hence extremists can be identified and the results used in interactive data analysis environments, such as software for exploratory network analysis. Applications include social network analysis, market basket analysis, epidemiology and molecular biology.