Integrated Node Encoder for Labelled Textual Networks

TL;DR

This paper introduces an integrated node encoder (INE) for textual networks that jointly learns from structure, text, and label information, improving classification accuracy and enabling label-enhanced embeddings for unlabelled nodes.

Contribution

The paper proposes a novel integrated node encoder that combines structure, text, and label data in a unified training framework, enhancing node embeddings and classification performance.

Findings

Achieved state-of-the-art classification accuracy on Cora and DBLP datasets.

Improved benchmark results by 10.0% and 12.1% with 70% training data.

Enables label-enhanced embeddings for unlabelled nodes.

Abstract

Voluminous works have been implemented to exploit content-enhanced network embedding models, with little focus on the labelled information of nodes. Although TriDNR leverages node labels by treating them as node attributes, it fails to enrich unlabelled node vectors with the labelled information, which leads to the weaker classification result on the test set in comparison to existing unsupervised textual network embedding models. In this study, we design an integrated node encoder (INE) for textual networks which is jointly trained on the structure-based and label-based objectives. As a result, the node encoder preserves the integrated knowledge of not only the network text and structure, but also the labelled information. Furthermore, INE allows the creation of label-enhanced vectors for unlabelled nodes by entering their node contents. Our node embedding achieves state-of-the-art performances in the classification task on two public citation networks, namely Cora and DBLP, pushing benchmarks up by 10.0\% and 12.1\%, respectively, with the 70\% training ratio. Additionally, a feasible solution that generalizes our model from textual networks to a broader range of networks is proposed.