Automatic depth-based local center clustering via $β$-integrated local depth and adaptive grouping
A new unsupervised clustering method, A-DLCC, estimates cluster count automatically without numerical parameter tuning.
Researchers propose automatic depth-based local center clustering (A-DLCC), an unsupervised method that avoids user-specified cluster counts and neighborhood sizes. It ranks stable exemplars with beta-integrated local depth and merges groups using a nonparametric similarity measure plus a bottleneck-path criterion checked against a configuration-model null. Experiments on synthetic and real data show the rule estimates the number of clusters and stops merging without numerical parameter tuning.
- A-DLCC clusters data without a preset cluster count or neighborhood size.
- Local centers are found with beta-integrated local depth.
- An adaptive merge rule estimates cluster count and when to stop.
- Synthetic and real-data tests require no numerical parameter tuning.
Full article201 words · extracted from arxiv.org · click to collapse
Clustering is an unsupervised learning technique that partitions unlabeled data into groups. Most existing methods require user-specified parameters, such as the number of clusters or neighborhood size. Conversely, we propose automatic depth-based local center clustering (A-DLCC), a fully data-driven method that eliminates numerical parameter tuning. A-DLCC uses the $β$-integrated local depth to identify stable exemplars, points consistently central across multiple locality levels, termed local centers, which are ranked by their representativeness. Each local center induces a group of similar points, with group-level similarity measured by a proposed nonparametric metric called group-level local similarity. To guide merging, we incorporate the bottleneck path idea from graph theory, which forms the basis of our adaptive merging criterion. Based on this criterion, we design a single agglomeration rule in which a group is either absorbed by a neighbor it reaches better than itself or bonded to a neighbor that both sides find more reachable than their own background, every merge being additionally required to be carried by a contact stronger than a configuration-model null expects. The rule automatically estimates the number of clusters and decides when to stop merging. Experiments on synthetic and real data show that A-DLCC produces interpretable clustering results without parameter tuning.
Text extracted automatically; images, tables and formatting may be missing. Original: https://arxiv.org/abs/2609.26748