An Artificial Bee Colony Optimization for Feature Subset Selection using Supervised Fuzzy C-Means Algorithm
Feature selection (FS) plays an important role in pattern recognition and machine learning systems. FS is used to improve the efficiency of learning algorithm especially for large scale datasets, by finding a minimal subset of features that has maximum efficacy on classifier. Artificial bee colony is a new optimization algorithm inspired of the natural behavior of honey bees in their search process for the best food sources. In this paper a new hybrid approach based on supervised fuzzy c-means algorithm and artificial bee colony optimization is proposed, that achieves an efficient feature subset selection. In the each iteration of artificial bee colony algorithm, the supervised fuzzy c-means clustering error rate and the length of selected feature subset vector are considered as heuristic information for the bees. Obtained results show that artificial bee colony algorithm has better results than the other optimization algorithm consist of genetic algorithm, particle swarm optimization, and ant colony optimization.
feature selection feature subset artificial bee colony optimization supervized fuzzy c-means algorithm
Mohammad Shokouhifar Fardad Farokhi
Member of Scientific Association of Electrical & Electronic Engineering Professor assistant of Electrical & Electronic Engineering Islamic Azad University of Central Tehran
国际会议
成都
英文
1702-1707
2010-12-17（万方平台首次上网日期，不代表论文的发表时间）