Concentric RadViz: Visual Exploration of Multi-task Classification

  title={Concentric RadViz: Visual Exploration of Multi-task Classification},
  author={Jorge Henrique Piazentin Ono and Fabio Sikansi and D{\'e}bora C. Corr{\^e}a and Fernando Vieira Paulovich and Afonso Paiva and Luis Gustavo Nonato},
  journal={2015 28th SIBGRAPI Conference on Graphics, Patterns and Images},
The discovery of patterns in large data collections is a difficult task. Visualization and machine learning techniques have emerged as a way to facilitate data analysis, providing tools to uncover relevant patterns from the data. This paper presents Concentric Rad Viz, a general purpose class visualization system that takes into account multi-class, multi-label and multi-task classifiers. Concentric Rad Viz uses a force attenuation scheme, which minimizes cluttering and ambiguity in the visual… CONTINUE READING


Publications referenced by this paper.

From music similarity to music recommendation: Computational approaches based on audio features and metadata

  • D. Bogdanov
  • Ph.D. dissertation, Universitat Pompeu Fabra…
  • 2013
2 Excerpts

Similar Papers

Loading similar papers…