Contrastive Multi-graph Learning with Neighbor Hierarchical Sifting for Semi-supervised Text Classification

TL;DR

This paper introduces ConNHS, a novel contrastive multi-graph learning method with neighbor hierarchical sifting for semi-supervised text classification, effectively leveraging multi-relational graphs and reducing false negatives to improve classification accuracy.

Contribution

The paper proposes a new contrastive learning framework that incorporates neighbor hierarchical sifting and relation-aware propagation to better utilize graph information and minimize false negatives in semi-supervised text classification.

Findings

Achieved over 95% accuracy on ThuCNews dataset.

Demonstrated superior performance on multiple datasets compared to existing methods.

Effectively reduces false negatives in contrastive learning, improving embedding quality.

Abstract

Graph contrastive learning has been successfully applied in text classification due to its remarkable ability for self-supervised node representation learning. However, explicit graph augmentations may lead to a loss of semantics in the contrastive views. Secondly, existing methods tend to overlook edge features and the varying significance of node features during multi-graph learning. Moreover, the contrastive loss suffer from false negatives. To address these limitations, we propose a novel method of contrastive multi-graph learning with neighbor hierarchical sifting for semi-supervised text classification, namely ConNHS. Specifically, we exploit core features to form a multi-relational text graph, enhancing semantic connections among texts. By separating text graphs, we provide diverse views for contrastive learning. Our approach ensures optimal preservation of the graph information, minimizing data loss and distortion. Then, we separately execute relation-aware propagation and cross-graph attention propagation, which effectively leverages the varying correlations between nodes and edge features while harmonising the information fusion across graphs. Subsequently, we present the neighbor hierarchical sifting loss (NHS) to refine the negative selection. For one thing, following the homophily assumption, NHS masks first-order neighbors of the anchor and positives from being negatives. For another, NHS excludes the high-order neighbors analogous to the anchor based on their similarities. Consequently, it effectively reduces the occurrence of false negatives, preventing the expansion of the distance between similar samples in the embedding space. Our experiments on ThuCNews, SogouNews, 20 Newsgroups, and Ohsumed datasets achieved 95.86\%, 97.52\%, 87.43\%, and 70.65\%, which demonstrates competitive results in semi-supervised text classification.