MSGKA: an efficient clustering algorithm for large databases

This investigation presents an efficient clustering algorithm for large databases. We present a novel multiple-searching genetic algorithm (MSGA) that finds a globally optimal partition of a given data into a specified number of clusters. We hybridize MSGA with a multiple-searching approach utilized in clustering namely, K-means algorithm. Hence, the name multiple-searching genetic K-means algorithm (MSGKA). Our simulation results reveal that the proposed novel clustering approach performs better than the Fast SOM combines K-means approach (FSOM+K-means) and Genetic K-Means Algorithm (GKA). Moreover, in all the cases we studied, our approach produces much smaller errors than both the FSOM+K-means and GKA.