Index-Based Solutions for Efficient Density Peak Clustering

Density Peak Clustering (DPC), a popular density-based clustering approach, has received considerable attention from the research community primarily due to its simplicity and fewer-parameter requirement. However, resultant clusters obtained using DPC are influenced by sensitive parameter $d_c$ , which depends on data distribution requirements of different users. Besides, original algorithm requires visiting large number objects, making it slow. To this end, paper investigates index-based solutions for DPC. Specifically, we propose two list-based index methods viz. (i) simple List Index, (ii) an advanced Cumulative Histogram Index. Efficient query algorithms proposed these indices significantly avoids irrelevant comparisons at cost space. For memory-constrained systems, further introduce approximate solution above allows substantial reduction in space cost, provided that slight inaccuracies admissible. Furthermore, owing considerably lower memory existing tree-based structures, also present effective pruning techniques efficient support Quadtree Index R-tree Finally, practically evaluate all findings results, set extensive experiments six synthetic real datasets. The experimental insights can help guide selecting befitting index.