Advanced search
Start date
Betweenand

Interactive visualization tool for exploring and learning from ensemble of clustering solutions

Grant number: 15/21560-0
Support type:Scholarships abroad - Research
Effective date (Start): March 28, 2016
Effective date (End): March 27, 2017
Field of knowledge:Physical Sciences and Mathematics - Computer Science
Principal Investigator:Katti Faceli
Grantee:Katti Faceli
Host: Julia Handl
Home Institution: Centro de Ciências em Gestão e Tecnologia (CCGT). Universidade Federal de São Carlos (UFSCAR). Campus de Sorocaba. Sorocaba , SP, Brazil
Local de pesquisa : University of Manchester, England  

Abstract

Cluster analysis has been largely employed for knowledge extraction in several fields. Interpretation of clustering results are not straightforward for a domain expert and visual aids are extremely useful in these cases. Techniques for visualization and full exploration of advanced clustering techniques results are still needed. Partitions' Visualizer (PVis) is a visualization tool that provides a static view of a collection of partitions, aiming at facilitating the work of domain experts comparing several clustering results. This research project aims to design/develop a visualization tool suitable for interactively explore collections of clustering solutions. The tool will take into account the capabilities of PVis, mechanisms for interaction, integration with recent clustering resources and integration with other already existing visualization devices. In order to show the effectiveness of the proposed tool, we will perform a case study. It will concern exploring the relations between partitions obtained with multi-objective clustering algorithms in the context of bioinformatics.