Concentric RadViz: Visual Exploration of Multi-task Classification

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 layout. In~addition, the user can add concentric circles to the layout in order to represent classification tasks. Our validation results and the application of Concentric Rad Viz for two real collections suggest that this tool can reveal important data patterns and relations. In our application, the user can interact with the visualization by selecting regions of interest according to specific criteria and changing projection parameters.