A new automatic density cluster

Clustering is an important unsupervised learning approach with wide application in data mining, pattern recognition and intelligent information processing. Density clustering is a classical clustering method for handling with non-spherical clusters, besides density clustering could operate by itself without setting the number of clusters, so it is extensive used by us. But there are some avoidless defects, such as, for asymmetry density data the effect is unsatisfactory. This paper introduce a new automatic density cluster algorithm, it brings the neighborhood radius on behalf of the density, then it abandons “minPts”, so change the way to grabble clusters, the Clustering Results have a phanerous improvement.