A new approach to cluster analysis

• Grant number: 07/00167-1
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: April 01, 2007
• End date: January 28, 2008
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: André Carlos Ponce de Leon Ferreira de Carvalho
• Grantee: Katti Faceli
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The traditional clustering techniques presents several restrictions and difficulties. Most of them are overcome by approaches such as clustering ensembles, multi-objective clustering (MOCK) and multi-objective ensemble (MOCLE). However, one dificultie still remains for the data having clusters with different shapes in different regions of the feature space. In this project, it will be proposed the development of a technique to the obtaining of independent clusters, in accordance with different clustering criteria, that lead to the most complete description of the data that is possible. The idea is that all (or most of) relevant clusters could be found, regardless the refinement level they occur, the clustering criteria they is in accordance with, or if the clusters are superimposed.