Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HACD: Harnessing Attribute Semantics and Mesoscopic Structure for Community Detection

Anran Zhang, Xingfen Wang, Yuhan Zhao

TL;DR

HACD is proposed, a novel attributed community detection model based on heterogeneous graph attention networks that treats node attributes as another type of node, constructs attributed networks into heterogeneous graph structures and employs attribute-level attention mechanisms to capture semantic similarity.

Abstract

Community detection plays a pivotal role in uncovering closely connected subgraphs, aiding various real-world applications such as recommendation systems and anomaly detection. With the surge of rich information available for entities in real-world networks, the community detection problem in attributed networks has attracted widespread attention. While previous research has effectively leveraged network topology and attribute information for attributed community detection, these methods overlook two critical issues: (i) the semantic similarity between node attributes within the community, and (ii) the inherent mesoscopic structure, which differs from the pairwise connections of the micro-structure. To address these limitations, we propose HACD, a novel attributed community detection model based on heterogeneous graph attention networks. HACD treats node attributes as another type of node, constructs attributed networks into heterogeneous graph structures and employs attribute-level attention mechanisms to capture semantic similarity. Furthermore, HACD introduces a community membership function to explore mesoscopic community structures, enhancing the robustness of detected communities. Extensive experiments demonstrate the effectiveness and efficiency of HACD, outperforming state-of-the-art methods in attributed community detection tasks. Our code is publicly available at https://github.com/Anniran1/HACD1-wsdm.

HACD: Harnessing Attribute Semantics and Mesoscopic Structure for Community Detection

TL;DR

HACD is proposed, a novel attributed community detection model based on heterogeneous graph attention networks that treats node attributes as another type of node, constructs attributed networks into heterogeneous graph structures and employs attribute-level attention mechanisms to capture semantic similarity.

Abstract

Community detection plays a pivotal role in uncovering closely connected subgraphs, aiding various real-world applications such as recommendation systems and anomaly detection. With the surge of rich information available for entities in real-world networks, the community detection problem in attributed networks has attracted widespread attention. While previous research has effectively leveraged network topology and attribute information for attributed community detection, these methods overlook two critical issues: (i) the semantic similarity between node attributes within the community, and (ii) the inherent mesoscopic structure, which differs from the pairwise connections of the micro-structure. To address these limitations, we propose HACD, a novel attributed community detection model based on heterogeneous graph attention networks. HACD treats node attributes as another type of node, constructs attributed networks into heterogeneous graph structures and employs attribute-level attention mechanisms to capture semantic similarity. Furthermore, HACD introduces a community membership function to explore mesoscopic community structures, enhancing the robustness of detected communities. Extensive experiments demonstrate the effectiveness and efficiency of HACD, outperforming state-of-the-art methods in attributed community detection tasks. Our code is publicly available at https://github.com/Anniran1/HACD1-wsdm.

Paper Structure

This paper contains 29 sections, 22 equations, 5 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Problem Statement
  4. Attributed Network
  5. HACD-Problem
  6. Attribute cohesiveness.
  7. Structure cohesiveness.
  8. Methodology
  9. Overall Framework
  10. Graph Constructing and Encoding
  11. Attribute-level Attention Mechanism
  12. Community Membership Function
  13. Training
  14. Experiments
  15. Experimental Setup
  16. ...and 14 more sections

Figures (5)

  • Figure 1: Most studies treat AI (artificial intelligence), CV (computer vision), and ML (machine learning) as independent attributes. However, AI and CV are subfields within the broader domain of ML, implying that they share underlying semantic similarities.
  • Figure 2: The overall framework of HACD.
  • Figure 3: The impact of the original graph structure and the updated graph structure for model performance.
  • Figure 4: The impact of parameters on HACD on the Cora dataset.
  • Figure 5: The robustness and scalability of HACD on the DBLP dataset.