Skip navigation

Meerkat

Social Network Analysis Tool

 

Principal Investigators Dr. Osmar Zaiane
zaiane@cs.ualberta.ca		 Dr. Randy Goebel
goebel@cs.ualberta.ca
Social networks are ubiquitous. Internet social networking is what most people know about but the analysis of social networks is important in many domains. Biologists study protein interactions forming a network. Criminologists and law enforcement agencies analyze crime networks. Epidemiologists study relationships between individuals. Zoologists examine animal behaviours materialized in networks. Telecommunication researchers investigate call networks. When these networks are small, the manual analysis is tractable but with large networks it becomes impossible. Social network analysis attempts to answer questions such as: which elements are the most influential? Which individuals are leaders and which ones are followers. Are there groups and how are they formed? Which elements within a group are important? What are the outliers? Which relationships are important?


Meerkat is a tool developed by a team lead by Osmar Zaïane and Randy Goebel at the Alberta Ingenuity Centre for Machine Learning to help practitioners better comprehend and analyze vast amount of data in the form of social networks. Given a set of objects and their interaction with each other, Meerkat uses sophisticated algorithms, developed at the Alberta Ingenuity Centre for Machine Learning, to automatically identify groups of objects that are meaningfully connected, that we call communities. Meerkat carefully lays the network out on the screen to minimize occlusion and highlights communities. It also provides information about the most influential/central nodes within each community.

Features:

  • Computes different measures of centrality
  • Provides different layouts
  • Automatically detects communities
  • Shows community dynamics in time