COMLApr 8, 2019

CRAD: Clustering with Robust Autocuts and Depth

arXiv:1904.04020v111 citations
Originality Incremental advance
AI Analysis

This work addresses clustering challenges for data with varying densities, offering incremental improvements in robustness and applicability to time series.

The authors tackled the problem of detecting clusters with varying densities by developing CRAD, a new density-based clustering algorithm that uses robust data depth as a dissimilarity measure, and showed it is highly competitive with existing methods like DBSCAN, OPTICS, and DBCA. They also introduced a parameter selection procedure for real-world clustering without ground truth and extended CRAD to time series clustering without prior knowledge of cluster numbers.

We develop a new density-based clustering algorithm named CRAD which is based on a new neighbor searching function with a robust data depth as the dissimilarity measure. Our experiments prove that the new CRAD is highly competitive at detecting clusters with varying densities, compared with the existing algorithms such as DBSCAN, OPTICS and DBCA. Furthermore, a new effective parameter selection procedure is developed to select the optimal underlying parameter in the real-world clustering, when the ground truth is unknown. Lastly, we suggest a new clustering framework that extends CRAD from spatial data clustering to time series clustering without a-priori knowledge of the true number of clusters. The performance of CRAD is evaluated through extensive experimental studies.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes