Density Adaptive Parallel Clustering

TL;DR

This paper introduces a new density-based clustering algorithm inspired by DBSCAN and minimum spanning trees, which is deterministic, simpler, faster, and does not require pre-setting the number of clusters.

Contribution

The proposed method offers a novel density-adaptive clustering approach that improves simplicity and speed without needing to specify cluster count beforehand.

Findings

The new algorithm is faster than previous solutions.

It is deterministic and does not require pre-setting k.

It compares favorably with existing density-based methods.

Abstract

In this paper we are going to introduce a new nearest neighbours based approach to clustering, and compare it with previous solutions; the resulting algorithm, which takes inspiration from both DBscan and minimum spanning tree approaches, is deterministic but proves simpler, faster and doesnt require to set in advance a value for k, the number of clusters.