BiTe-GCN: A New GCN Architecture via BidirectionalConvolution of Topology and Features on Text-Rich Networks

TL;DR

BiTe-GCN introduces a bidirectional convolutional architecture that simultaneously processes topology and textual features in text-rich networks, overcoming limitations of traditional GCNs like over-smoothing and local homophily, leading to superior performance.

Contribution

The paper presents a novel GCN architecture with bidirectional convolution on both topology and features, specifically designed for text-rich networks, and demonstrates its effectiveness through extensive experiments.

Findings

Outperforms state-of-the-art GCN models on text-rich networks

Effective in e-commerce search scenarios like JD search

Addresses over-smoothing and topology-feature imbalance issues

Abstract

Graph convolutional networks (GCNs), aiming to integrate high-order neighborhood information through stacked graph convolution layers, have demonstrated remarkable power in many network analysis tasks. However, topological limitations, including over-smoothing and local topology homophily, limit its capability to represent networks. Existing studies only perform feature convolution on network topology, which inevitably introduces unbalance between topology and features. Considering that in real world, the information network consists of not only the node-level citation information but also the local text-sequence information. We propose BiTe-GCN, a novel GCN architecture with bidirectional convolution of both topology and features on text-rich networks to solve these limitations. We first transform the original text-rich network into an augmented bi-typed heterogeneous network, capturing both the global node-level information and the local text-sequence information from texts. We then introduce discriminative convolution mechanisms to performs convolutions of both topology and features simultaneously. Extensive experiments on text-rich networks demonstrate that our new architecture outperforms state-of-the-art by a breakout improvement. Moreover, this architecture can also be applied to several e-commerce searching scenes such as JD searching. The experiments on the JD dataset validate the superiority of the proposed architecture over the related methods.