Citation network applications in a scientific co-authorship recommender system

TL;DR

This paper introduces a novel co-authorship recommendation system that leverages citation network data and graph neural networks to predict potential scientific collaborations effectively.

Contribution

It presents a new pipeline utilizing citation data and graph neural networks for link prediction in co-authorship networks, enhancing collaboration recommendations.

Findings

Effective use of citation data improves link prediction accuracy.

Graph neural networks outperform traditional methods in co-authorship prediction.

The proposed system demonstrates promising results in recommending potential collaborators.

Abstract

The problem of co-authors selection in the area of scientific collaborations might be a daunting one. In this paper, we propose a new pipeline that effectively utilizes citation data in the link prediction task on the co-authorship network. In particular, we explore the capabilities of a recommender system based on data aggregation strategies on different graphs. Since graph neural networks proved their efficiency on a wide range of tasks related to recommendation systems, we leverage them as a relevant method for the forecasting of potential collaborations in the scientific community.