• TUCS Publication Series
• How to Publish in TUCS Series
• Publication Search
• Publication Graph
• Publication Input
• Publication Input Guide
• JuFo Browser

Interactive Visual Exploration of Topic Models Using Graphs

Samuel Rönnqvist, Xiaolu Wang, Peter Sarlin, Interactive Visual Exploration of Topic Models Using Graphs. In: Robert S Laramee, Min Chen (Eds.), The Eurographics Conference on Visualization (EuroVis), Eurographics, 2014.

Abstract:

Probabilistic topic modeling is a popular and powerful family of tools for uncovering thematic structure in large sets of unstructured text documents. The extensive research into this type of modeling has meet comparatively few studies concerning how to present or visualize topic models in meaningful ways. In this paper, we present a design that uses graphs to visually communicate topic structure and meaning, as uncovered by unsupervised modeling. By connecting topic nodes via descriptive keyterms, the graph representation reveals topic similarities, topic meaning and shared, ambiguous keyterms, while also supporting information retrieval by topic or topic subsets.

BibTeX entry:

@INPROCEEDINGS{inpRxWaSa14a,
  title = {Interactive Visual Exploration of Topic Models Using Graphs},
  booktitle = {The Eurographics Conference on Visualization (EuroVis)},
  author = {Rönnqvist, Samuel and Wang, Xiaolu and Sarlin, Peter},
  editor = {Laramee, Robert S and Chen, Min},
  publisher = {Eurographics},
  year = {2014},
}

Belongs to TUCS Research Unit(s): Data Mining and Knowledge Management Laboratory

Publication Forum rating of this publication: level 1

Edit publication