Visualization of network data for effective semantic analysis
- Resource Type
- Conference
- Authors
- Krishna, Varun; Suri, N N R Ranga; Kumar, K R Prasanna; Rakshit, Subrata
- Source
- 2015 IEEE Workshop on Computational Intelligence: Theories, Applications and Future Directions (WCI) Computational Intelligence: Theories, Applications and Future Directions (WCI), 2015 IEEE Workshop on. :1-6 Dec, 2015
- Subject
- Computing and Processing
Power, Energy and Industry Applications
Signal Processing and Analysis
Data visualization
Visualization
Semantics
Image color analysis
Data mining
Java
Social network services
Data Visualization
Network Data
Semantic analysis
Visual Analytics
Graph mining
- Language
Visual data analysis has been emerging as an attractive technique in addressing the challenges of the current computing research. It relies upon the human ability in grasping the intrinsic characteristics of the data through visual exploration. This human visual perception combined with the domain specific data semantics can lead to effective extraction of knowledge that is of immense use in real life. However, visualization of network data poses some serious issues as it has to deal with the inherent links between various network entities in addition to rendering the entities themselves in a semantically appealing manner. Addressing these issues, this paper proposes a methodology of employing graph-based representation of network data for extracting useful semantic graph patterns from the data. Our method defines various visual features that are used in rendering an interactive visual display of the network data for carrying out semantic analysis. An implementation of the proposed method is achieved by extending the Java Universal Network/Graph (JUNG) library. The efficacy of the proposed method is demonstrated through an experimental study using some benchmark network data sets.