Hybrid data clustering approaches using bacterial colony optimization and k-means

Abstract

Data clustering is a fashionable data analysis technique in the data mining. K-means is a popular clustering technique for solving a clustering problem. However, the k-means clustering technique extremely depends on the initial position and converges to a local optimum. On the other hand, the bacterial colony optimization (BCO) is a well-known recently proposed data clustering algorithm. However, it is a high computational cost to complete a given solution. Hence, this research paper proposes a new hybrid data clustering method for solving data clustering problem. The proposed hybrid data clustering algorithm is a combination of the BCO and K-means called BCO+KM clustering algorithm. The experimental result shows that the proposed hybrid BCO+KM data clustering algorithm reveal better cluster partitions.