A Comprehensive Approach to Mode Clustering

Abstract

Mode clustering is a nonparametric method for clustering that defines clusters using the basins of attraction of a density estimator's modes. We provide several enhancements to mode clustering: (i) a soft variant of cluster assignment, (ii) a measure of connectivity between clusters, (iii) a technique for choosing the bandwidth, (iv) a method for denoising small clusters, and (v) an approach to visualizing the clusters. Combining all these enhancements gives us a complete procedure for clustering in multivariate problems. We also compare mode clustering to other clustering methods in several examples

34 pages, 17 figures. Accepted to the Electronic Journal of Statistics. The original title is "Enhanced Mode Clustering"

References in corpus (6)

• Natural Scales in Geographical Patterns
• Nonparametric modal regression
• Robust Topological Inference: Distance To a Measure and Kernel Distance
• Pruning nearest neighbor cluster trees
• Generalized Mode and Ridge Estimation
• Clusters and water flows: a novel approach to modal clustering through Morse theory

Cited by in corpus (3)

• Nonparametric modal regression
• Density Level Sets: Asymptotics, Inference, and Visualization
• Sparse Approximation of a Kernel Mean