An Improved Clustering Algorithm Based on Ant Colony Approach

Abstract

Ant colony algorithm is a kind of evolutionary algorithm with global optimization quality to deal with discrete problem. Clustering analysis is an important part in data mining community. Traditional clustering algorithm is slow of the convergence and sensitive to the initial value and preset classed in large scale data set. The ant colony algorithm was applied in aggregation analysis for the first time in this paper. A new clustering algorithm was presented based on the ant colony algorithm. This algorithm has quality of essential parallel, quick convergence and high effectiveness. The experimental result shows that it is about 10% higher than the c-means method in effectiveness.