NodeTrix-Multiplex: Visual Analytics of Multiplex Small World Networks

Year: 2017

Conference: Complex Networks & Their Applications

Authors: Shivam Agarwal, Amit Tomar, and Jaya Sreevalsan-Nair

DOI: 10.1007/978-3-319-50901-3_46

Abstract

Analyzing multiplex small world networks (SWNs) using community detection (CD) is a challenging task. We propose the use of visual analytics to probe and extract communities in such networks, where one of the layers defines the network topology and exhibits small-world property. Our novel visual analytics framework, NodeTrix-Multiplex (NTM), for visual exploration of multiplex SWNs, integrates focus+context network visualization, and analysis of community detection results, within the focus. We propose a heterogeneous data model, which composites multiple layers for the focus and context and thus, enables finding communities across layers. We perform a case-study on a co-authorship (collaboration) network, with a functional layer obtained from the author-topic similarity graph. We also perform an expert user evaluation of the tool, developed using NTM.

Citation

@inproceedings{Agarwal2017Nodetrix,
  author     = {Shivam Agarwal, Amit Tomar, and Jaya Sreevalsan-Nair},
  title      = {NodeTrix-Multiplex: Visual Analytics of Multiplex Small World Networks},
  booktitle  = {Complex Networks & Their Applications},
  pages      = {579--591},
  year       = {2017},
  publisher  = {Springer International Publishing},
  doi        = {10.1007/978-3-319-50901-3_46},
  paperurl   = {/publications/Agarwal2017Nodetrix/Agarwal2017Nodetrix.pdf},
  demo       = {https://nodetrix-multiplex.vercel.app/},
  abstract   = {Analyzing multiplex small world networks (SWNs) using community detection (CD) is a challenging task. We propose the use of visual analytics to probe and extract communities in such networks, where one of the layers defines the network topology and exhibits small-world property. Our novel visual analytics framework, NodeTrix-Multiplex (NTM), for visual exploration of multiplex SWNs, integrates focus+context network visualization, and analysis of community detection results, within the focus. We propose a heterogeneous data model, which composites multiple layers for the focus and context and thus, enables finding communities across layers. We perform a case-study on a co-authorship (collaboration) network, with a functional layer obtained from the author-topic similarity graph. We also perform an expert user evaluation of the tool, developed using NTM.}
}