fast-ldp-mst-clustering 1.0.1

Fast_LDP_MST is an efficient density-based clustering method for large-size datasets.

fast_ldp_mst_clustering

Fast LDP-MST is an efficient density-based clustering method for large-size datasets.

Teng Qiu, Yongjie Li, IEEE Transactions on Knowledge and Data Engineering, 2022, DOI: 10.1109/TKDE.2022.3150403.

Cluster analysis is widely studied and used in diverse fields. Despite that many clustering methods have been proposed, it is rare for a clustering method to perform well simultaneously on the following characteristics:

• effectiveness (in terms of accuracy);
• efficiency (in terms of speed);
• robustness (in terms of noise and parameter sensitivity);
• user-friendliness (in terms of the number of user-specified parameters, interpretability, and reproducibility of the results, etc.).

This work contributes to the field of clustering by providing a new solution (i.e., Fast LDP-MST) which largely improves the efficiency of LDP-MST, without sacrificing the merits of LDP-MST in effectiveness, robustness, and user-friendliness. Fast LDP-MST achieves a good balance among effectiveness, efficiency, robustness, and user-friendliness, and thus it could have a certain degree of practical value in this big data era.

The implementation of the MATLAB version can be accessed through the link: https://github.com/Teng-Qiu-Clustering/Fast-LDP-MST-Clustering-IEEE-TKDE-2022