SCHOLAT Link Prediction: A Link Prediction Dataset Fusing Topology and Attribute Information

Ronghua Lin, Yong Tang, Chengzhe Yuan, Chaobo He, Weisheng Li

Abstract: Link prediction is an important research field on social network analysis. However, most existing link prediction datasets have not taken text attribute information into account. In this paper, we propose a novel link prediction dataset named SCHOLAT Link Prediction based on the academic social network SCHOLAT, which contains multitype relationships and user attribute information. We conduct extensive experiments on our dataset and the results show the potential of our dataset on link prediction. Our proposed dataset is available at https://www.scholat.com/research/opendata/#link_prediction.