A New Clustering Algorithm Based on Near Neighbor Influence

TL;DR

This paper introduces CNNI, a novel clustering algorithm inspired by near neighbor influence and superposition principles, demonstrating good performance on artificial and real datasets with proper parameter tuning.

Contribution

It proposes a new clustering method based on near neighbor influence, including new concepts and similarity measures, with experimental validation.

Findings

Effective clustering on artificial datasets

Good performance on real datasets with proper parameters

Introduces new concepts like near neighbor influence

Abstract

This paper presents Clustering based on Near Neighbor Influence (CNNI), a new clustering algorithm which is inspired by the idea of near neighbor and the superposition principle of influence. In order to clearly describe this algorithm, it introduces some important concepts, such as near neighbor point set, near neighbor influence, and similarity measure. By simulated experiments of some artificial data sets and seven real data sets, we observe that this algorithm can often get good clustering quality when making proper value of some parameters. At last, it gives some research expectations to popularize this algorithm.