An Enhanced Density Based Spatial clustering of Applications with Noise

Cluster analysis is a primary method for data mining. Finding clusters with varying sizes, shapes and densities is a challenging job. DBSCAN can find clusters with varying shapes and sizes. But it has a trouble in finding clusters with varying densities, because it depends on a global value for its parameter Eps. This paper presents enhanced DBSCAN which clusters databases containing clusters with varying densities effectively. The idea is to use varied values for Eps according to the local density of the starting point in each cluster. The clustering process starts from the highest local density point towards the lowest local density one. For each value of Eps, DBSCAN is adopted to make sure that all density reachable points with respect to current Eps are clustered. At the next process, the clustered points are ignored, to avoid merging among denser clusters with sparser ones.