Clustering Methods Based on Closest String via Rank Distance

This paper aims to present two clustering methods based on rank distance. Rank distance has applications in many different fields such as computational linguistics, biology and informatics. Rank distance can be computed fast and benefits from some features of the edit (Levenshtein) distance. In [1] two clustering methods based on rank distance are described. The K-means algorithm uses the median string to represent the centroid of a cluster, while the hierarchical clustering method joins pairs of strings and replaces each pair with the median string. Two similar clustering algorithms are about to be presented in this paper, only that the closest string will be considered instead of the median string. The new clustering algorithms are compared with those presented in [1] and other similar clustering techniques. Experiments using mitochondrial DNA sequences extracted from several mammals are performed to compare the results of the clustering methods. Results demonstrate the clustering performance and the utility of the new algorithms.