Date Added: Nov 2011
Clustering is a splitting up of data into groups of similar objects called clusters. The objects in a cluster are similar between themselves and dissimilar compared to objects of other clusters. This paper is intended to study and compare different data clustering algorithms. The algorithms in investigation are: hierarchical agglomerative clustering algorithms: Parameter Free Minimum Spanning Tree (MST) clustering algorithm and single link, complete link and average link clustering algorithms. K-means partitional clustering algorithm is used in the results as a reference. The authors' experimental evaluation shows that Parameter Free Minimum Spanning Tree algorithms are lead to better clustering results than hierarchical agglomerative algorithms, which suggests that Parameter Free Minimum Spanning Tree clustering algorithms are well-suited for clustering.