A Comparative Analysis of Density Based Clustering Techniques for Outlier Mining

Density based Clustering Algorithms such as Density Based Spatial Clustering of Applications with Noise (DBSCAN), Ordering Points to Identify the Clustering Structure (OPTICS) and DENsity based CLUstering (DENCLUE) are designed to discover clusters of arbitrary shape. DBSCAN grows clusters according to a density based connectivity analysis. OPTICS, which is an extension of DBSCAN used to produce clusters ordering obtained by setting range of parameter. DENCLUE clusters object is based on a set of density distribution functions. The comparison of the algorithms in terms of essential parameters such as complexity, clusters shape, input parameters, noise handle, cluster quality and run time are considered. The analysis is useful in finding which density based clustering algorithm is suitable in different criteria.