A text mining model based on improved density clustering algorithm

Abstract

The clustering algorithm based on density is widely used on text mining model, for example the DBSCAN(density-based spatial clustering of application with noise) algorithm. DBSCAN algorithm is sensitive in choose of parameters, it is hard to find suitable parameters. In this paper a method based on k-means algorithm is introduced to estimate the e neighborhood and minpts. Finally an example is given to show the effectiveness of this algorithm.