On the Utilization of Structural and Textual Information of a Scientific Knowledge Graph to Discover Future Research Collaborations: A Link Prediction Perspective

Abstract

We consider the discovery of future research collaborations as a link prediction problem applied on scientific knowledge graphs. Our approach integrates into a single knowledge graph both structured and unstructured textual data through a novel representation of multiple scientific documents. The Neo4j graph database is used for the representation of the proposed scientific knowledge graph. For the implementation of our approach, we use the Python programming language and the scikit-learn ML library. We benchmark our approach against classical link prediction algorithms using accuracy, recall, and precision as our performance metrics. Our initial experimentations demonstrate a significant improvement of the accuracy of the future collaboration prediction task. The experimentations reported in this paper use the new COVID-19 Open Research Dataset.