Hierarchical Clustering Algorithm Based on Neighborhood-Linked in Large Spatial Databases

Abstract

A novel hierarchical clustering algorithm based on neighborhood-linked is proposed in this paper. Unlike the traditional hierarchical clustering algorithm, the new model only adopts two steps: clustering primarily and merging. The algorithm can be performed in high-dimensional data set, clustering the arbitrary shape of clusters. Furthermore, not only can this algorithm dispose the data with numeric attributes, but with boolean and categorical attributes. The results of our experimental study in data sets with arbitrary shape and size are very encouraging. We also conduct an experimental study with web log files that can help us to discover the use access patterns effectively. Our study shows that this algorithm generates better quality clusters than traditional algorithms, and scales well for large spatial databases.