Graph Embedding for Mapping Interdisciplinary Research Networks

TL;DR

This paper evaluates various research paper embedding methods for interdisciplinary citation prediction and introduces a novel Graph Neural Network architecture that better preserves interdisciplinary information in citation networks.

Contribution

It proposes a new GNN architecture tailored to maintain interdisciplinary insights in research paper embeddings, outperforming existing GNN methods in citation prediction tasks.

Findings

The proposed GNN outperforms other methods in interdisciplinary citation prediction.

The new architecture maintains overall citation prediction performance.

Evaluation on citation prediction tasks demonstrates the effectiveness of the approach.

Abstract

Representation learning is the first step in automating tasks such as research paper recommendation, classification, and retrieval. Due to the accelerating rate of research publication, together with the recognised benefits of interdisciplinary research, systems that facilitate researchers in discovering and understanding relevant works from beyond their immediate school of knowledge are vital. This work explores different methods of research paper representation (or document embedding), to identify those methods that are capable of preserving the interdisciplinary implications of research papers in their embeddings. In addition to evaluating state of the art methods of document embedding in a interdisciplinary citation prediction task, we propose a novel Graph Neural Network architecture designed to preserve the key interdisciplinary implications of research articles in citation network node embeddings. Our proposed method outperforms other GNN-based methods in interdisciplinary citation prediction, without compromising overall citation prediction performance.