OPTICS: ordering points to identify the clustering structure
SIGMOD '99 Proceedings of the 1999 ACM SIGMOD international conference on Management of data
G-TREACLE: a new grid-based and tree-alike pattern clustering technique for large databases
PAKDD'08 Proceedings of the 12th Pacific-Asia conference on Advances in knowledge discovery and data mining
Hi-index | 0.00 |
Data clustering plays an important role in various fields. Data clustering approaches have been presented in recent decades. Identifying clusters with widely differing shapes, sizes and densities in the presence of noise and outliers is challenging. Many density-based clustering algorithms, such as DBSCAN, can locate arbitrary shapes, sizes and filter noise, but cannot identify clusters based on differences in densities. Although the DD-DBSCAN algorithm can detect clusters with different densities, its time cost is considerable. This work presents a novel clustering method called DDCT (Density Difference by Clustering Technique), which is an extension of the existing DD-DBSCAN approach. The simulation reveals that the proposed approach derives clusters widely differing shapes, sizes and noise. Moreover, the proposed method can locate clusters with different densities rapidly. DDCT outperforms DD-DBSCAN.