An Effective Algorithm Based on Density Clustering Framework
Clustering analysis has the very broad applications on data analysis, such as data mining, machine learning, and information retrieval. In practice, most of clustering algorithms suffer from the effects of noises, different densities and shapes, cluster overlaps, etc. To solve the problems, in this paper, we propose a simple but effective density-based clustering framework (DCF) and implement a clustering algorithm based on DCF. In DCF, a raw data set is partitioned into core points and non-core points by a neighborhood density estimation model, and then the core points are clustered first, because they usually represent the center or dense region of the cluster structure. Finally, DCF classifies the non-core points into initial clusters in sequence. In experiments, we compare our algorithm with Dp and DBSCAN algorithms on synthetic and real-world data sets. The experimental results show that the performance of the proposed clustering algorithm is comparable with DBSCAN and Dp algorithms.
Clustering algorithms, reverse k-nearest neighbors, neighborhood density estimation, data mining, minimum spanning tree.