Linear Clustering Process on Networks

TL;DR

This paper introduces a novel linear clustering process for networks that uses attraction and repulsion forces based on neighborhood similarity to identify clusters, outperforming traditional modularity-based methods like Louvain.

Contribution

It presents a new linear clustering method that estimates the number of clusters and memberships using a force-based model, with superior performance on synthetic networks.

Findings

Outperforms Louvain method in clustering accuracy.

Maintains comparable computational complexity.

Effective on synthetic benchmark networks.

Abstract

We propose a linear clustering process on a network consisting of two opposite forces: attraction and repulsion between adjacent nodes. Each node is mapped to a position on a one-dimensional line. The attraction and repulsion forces move the nodal position on the line, depending on how similar or different the neighbourhoods of two adjacent nodes are. Based on each node position, the number of clusters in a network, together with each node's cluster membership, is estimated. The performance of the proposed linear clustering process is benchmarked on synthetic networks against widely accepted clustering algorithms such as modularity, the Louvain method and the non-back tracking matrix. The proposed linear clustering process outperforms the most popular modularity-based methods, such as the Louvain method, while possessing a comparable computational complexity.