Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Comparative Analysis of Community Detection Algorithms on the SNAP Social Circles Dataset

Yash Malode, Amit Aylani, Arvind Bhardwaj, Deepak Hajoary

TL;DR

This study conducts a systematic comparison of six prominent community detection algorithms on the SNAP Social Circles Facebook-derived dataset, using modularity, normalized cut, silhouette, compactness, Calinski-Harabasz, and separability as evaluation criteria. It implements K-Means, Louvain, Spectral Clustering, Label Propagation, Infomap, and Leading Eigenvector, reporting how many communities each method uncovers and how they score across metrics. The results highlight that Louvain and Label Propagation consistently perform well across multiple measures, though no single method is universally superior, underscoring the importance of task-specific algorithm selection. The work provides practical guidance for applying community detection to large real-world social networks and shares the accompanying code at the project repository for reproducibility.

Abstract

In network research, Community Detection has always been a topic of significant interest in network science, with numerous papers and algorithms proposing to uncover the underlying structures within networks. In this paper, we conduct a comparative analysis of several prominent community detection algorithms applied to the SNAP Social Circles Dataset, derived from the Facebook Social Media network. The algorithms implemented include Louvain, Girvan-Newman, Spectral Clustering, K-Means Clustering, etc. We evaluate the performance of these algorithms based on various metrics such as modularity, normalized cut-ratio, silhouette score, compactness, and separability. Our findings reveal insights into the effectiveness of each algorithm in detecting various meaningful communities within the social network, shedding light on their strength and limitations. This research contributes to the understanding of community detection methods and provides valuable guidance for their application in analyzing real-world social networks.

Comparative Analysis of Community Detection Algorithms on the SNAP Social Circles Dataset

TL;DR

This study conducts a systematic comparison of six prominent community detection algorithms on the SNAP Social Circles Facebook-derived dataset, using modularity, normalized cut, silhouette, compactness, Calinski-Harabasz, and separability as evaluation criteria. It implements K-Means, Louvain, Spectral Clustering, Label Propagation, Infomap, and Leading Eigenvector, reporting how many communities each method uncovers and how they score across metrics. The results highlight that Louvain and Label Propagation consistently perform well across multiple measures, though no single method is universally superior, underscoring the importance of task-specific algorithm selection. The work provides practical guidance for applying community detection to large real-world social networks and shares the accompanying code at the project repository for reproducibility.

Abstract

In network research, Community Detection has always been a topic of significant interest in network science, with numerous papers and algorithms proposing to uncover the underlying structures within networks. In this paper, we conduct a comparative analysis of several prominent community detection algorithms applied to the SNAP Social Circles Dataset, derived from the Facebook Social Media network. The algorithms implemented include Louvain, Girvan-Newman, Spectral Clustering, K-Means Clustering, etc. We evaluate the performance of these algorithms based on various metrics such as modularity, normalized cut-ratio, silhouette score, compactness, and separability. Our findings reveal insights into the effectiveness of each algorithm in detecting various meaningful communities within the social network, shedding light on their strength and limitations. This research contributes to the understanding of community detection methods and provides valuable guidance for their application in analyzing real-world social networks.

Paper Structure

This paper contains 38 sections, 10 equations, 18 figures.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Problem Statement
  4. Methodology
  5. Terminologies
  6. Network
  7. Node and Edges
  8. Community
  9. Degree
  10. Centrality
  11. Hub
  12. Path
  13. Cycle
  14. Connected Components
  15. Metrics
  16. ...and 23 more sections

Figures (18)

  • Figure 1: Approach Used for Implementation
  • Figure 2: Degree Distribution of nodes in the dataset
  • Figure 3: Cumulative Distribution Plot of nodes in the dataset
  • Figure 4: Degree Centrality Plot of each node
  • Figure 5:
  • ...and 13 more figures