Streamline density peak clustering for practical adoptions

Abstract

Since Density Peak Clustering (DPC) algorithm was proposed in 2014, it has drawn lots of interest in various domains. As a clustering method, DPC features superior generality, robustness, lexibility and simplicity. There are however two main roadblocks for its practical adoptions, both centered around the selection of cutof distance, the single critical hyperparameter of DPC. This work proposes an improved algorithm named Streamlined Density Peak Clustering (SDPC). SDPC speeds up DPC executions on a sequence of cutof distances by 2.2-8.8X while at the same time reducing memory usage by a magnitude. As an algorithm preserving the original semantic of DPC, SDPC ofers an eicient and scalable drop-in replacement of DPC for data clustering.