CaEGCN: Cross-Attention Fusion based Enhanced Graph Convolutional Network for Clustering

TL;DR

CaEGCN introduces a novel deep clustering framework that combines cross-attention fusion of content and relationship data with auto-encoders, improving clustering accuracy by leveraging heterogeneous data representations.

Contribution

The paper proposes CaEGCN, a new deep clustering method that integrates content and relationship information using cross-attention and auto-encoders, addressing over-smoothing in GCNs.

Findings

Outperforms existing clustering methods on various datasets.

Effectively fuses content and relationship data for improved clustering.

Demonstrates robustness across different data types.

Abstract

With the powerful learning ability of deep convolutional networks, deep clustering methods can extract the most discriminative information from individual data and produce more satisfactory clustering results. However, existing deep clustering methods usually ignore the relationship between the data. Fortunately, the graph convolutional network can handle such relationship, opening up a new research direction for deep clustering. In this paper, we propose a cross-attention based deep clustering framework, named Cross-Attention Fusion based Enhanced Graph Convolutional Network (CaEGCN), which contains four main modules: the cross-attention fusion module which innovatively concatenates the Content Auto-encoder module (CAE) relating to the individual data and Graph Convolutional Auto-encoder module (GAE) relating to the relationship between the data in a layer-by-layer manner, and the self-supervised model that highlights the discriminative information for clustering tasks. While the cross-attention fusion module fuses two kinds of heterogeneous representation, the CAE module supplements the content information for the GAE module, which avoids the over-smoothing problem of GCN. In the GAE module, two novel loss functions are proposed that reconstruct the content and relationship between the data, respectively. Finally, the self-supervised module constrains the distributions of the middle layer representations of CAE and GAE to be consistent. Experimental results on different types of datasets prove the superiority and robustness of the proposed CaEGCN.