A density-based clustering algorithm for weighted network with attribute information
Lingyu Wu, Xuedong Gao · 2011
Research of clustering method based on density is an important task in data mining. In order to improve the density-based methods for attribute space(such as DBSCAN, CLIQUE, OPTICS and so on) which ignore the relationships between objects, and the density-based methods for network(such as SCAN, DCSBRD and so on) which ignore the attribute information of objects, a density-based clustering algorithm for weighted network with attribute information (DCAWN) is proposed in the paper. After setting up the weighted network based on attribute distance, the algorithm refreshes the definition of near neighbor object and core object, and offers the corresponding clustering policy. For considering both attribute and relationship information, the algorithm increases the clustering accuracy, improves the clustering result, and distinguishes the hub and outlier objects effectively.