User:Im sankalp/sandbox

Meerkat (SNA) is a social network analysis software and GUI application developed in Java and JavaFx. It offers features for exploring social networks, visualising networks through multiple layout algorithms, analysing subsets of networks, performing community mining with several algorithms, and revealing network dynamics at the community level. It allows the user to compute descriptive node and community metrics and provides capabilities for event analysis for dynamic networks that have been observed at multiple points in time.

Meerkat was developed by a team lead by Osmar R. Zaiane at the Alberta Machine Intelligence Institute Amii to help practitioners better comprehend and analyse vast amounts of data in the form of social networks. Given a set of objects and their interactions with each other, Meerkat uses sophisticated algorithms, developed at Amii, to automatically identify groups of objects that are meaningfully connected, which we call communities. Meerkat carefully lays the network out on the screen to minimise occlusion and highlight communities. It also provides information about the most influential and central nodes within each community.

Features

 * Graph Editing via UI - Edit the structure of the graph via GUI interaction. The user can edit a graph by adding or deleting a set of vertices or edges.
 * Graph Canvas View - Manipulate views of the graph via features such as zoom in, zoom out and dragging the graph on the graph canvas area. The user can change the position of vertices on the graph by dragging them on the graph canvas area. The user can change the size and color of vertices and edges of the graph as well.
 * Minimap - Birds eye view of the graph. The minimap shows a snapshot of the entire graph where the graph display area is unshaded and the non-visible part is shaded.
 * Layouts - Meerkat provides three types of layout algorithms.
 * Standard Layouts - Circle, Random, Kamada-Kawai, Fruchterman-Reingold, ForceAtlas2 and Modified Fruchterman-Reingold
 * Metric Layouts - Authority, Hub, Betweenness, Centrality, Closeness, Centrality, In-Degree, Out-Degree and Pagerank. The result is plotted on graph canvas in bulls eye style.
 * Community Layouts - ComB and ComC


 * Community Mining - Compute and visualise groups of similar nodes based on edges in the graph. User can perform community mining by selecting any one of the following 6 algorithms in Meerkat : Fast Modularity, Louvain, K-means clustering, Local T,  Local Top Leaders, Rosvall Infomap, Rosvall Infomod, Same Attribute Value, Dynamic Community Mining and Local Community.
 * Metrics - User can calculate a number of graph metrics and measures to analyze the graph and its nodes. Authority, Hub, Betweenness, Centrality, Closeness, Centrality, In-Degree, Out-Degree and Pagerank metrics are provided in Meerkat.

Event Analysis
Evolution of communities using dynamic community mining. Meerkat provides algorithm to compute and analyse changes in the structure of communities over time. Critical events of communities and nodes can enable user to track the evolution of communities. This type of analysis provides insight into the underlying behaviour of the network.

There are nine different types of events in dynamic community mining which is divided into two different categories:


 * Community Events: Formed, Dissolved, Survived, Split and Merged
 * Node Events: Appeared, Disappeared, Joined and Left

File Formats
The various file formats supported by Meerkat are:
 * gml
 * graphml
 * pairs
 * meerkat
 * csv