Using ACO and rough set theory to feature selection

Authors:
Rafael Bello Pérez;Ann Nowe;Peter Vrancx;Yudel Gómez;Yailé D. Caballero
Affiliations:
Computer Science Department, Universidad Central de Las Villas, Santa Clara, Cuba;CoMo Lab, Computer Science Dept, Vrije Universiteit Brussel, Brussel, Belgium;CoMo Lab, Computer Science Dept, Vrije Universiteit Brussel, Brussel, Belgium;Computer Science Department, Universidad Central de Las Villas, Santa Clara, Cuba;Informatic Department, Universidad de Camaguey, Camaguey, Cuba
Venue:
EC'05 Proceedings of the 6th WSEAS international conference on Evolutionary computing
Year:
2005

Citing 8
Cited 0

Computational methods for rough classification and discovery

Journal of the American Society for Information Science - Special issue: knowledge discovery and data mining
Feature selection and effective classifiers

Journal of the American Society for Information Science - Special issue: knowledge discovery and data mining
Ant algorithms for discrete optimization

Artificial Life
Rough-Fuzzy Hybridization: A New Trend in Decision Making

Rough-Fuzzy Hybridization: A New Trend in Decision Making
Inclusion degree: a perspetive on measures for rough set data analysis

Information Sciences—Informatics and Computer Science: An International Journal
Consistency-based search in feature selection

Artificial Intelligence
Ant Colony Optimization

Ant Colony Optimization
The equation for response to selection and its use for prediction

Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we propose a model to feature selection based on ant colony and rough set theory (RST). The objective is to find the reducts. RST offers the heuristic function to measure the quality of one feature subset. We have studied three variants of ant's algorithms and the influence of the parameters on the performance both in terms of quality of the results and the number of reducts found. Experimental results show this hybrid approach shows interesting advantages when compared with other heuristic methods.