Scholar2vec: Vector Representation of Scholars for Lifetime Collaborator Prediction

While scientific collaboration is critical for a scholar, some collaborators can be more significant than others, e.g., lifetime collaborators. It has been shown that are influential on scholar’s academic performance. However, little research done investigating predicting such special relationships in networks. To this end, we propose Scholar2vec, novel neural network embedding representing scholar profiles. First, our approach creates scholars’ interest vector from textual information, as demographics, research, and influence. After bridging interests with network, representations of scholars gained graph learning. Meanwhile, since occupied various attributes, to incorporate four types attributes learning vectors. Finally, the early-stage similarity sequence based Scholar2vec used predict machine methods. Extensive experiments two real-world datasets show outperforms state-of-the-art methods collaborator prediction. Our work presents new way measure between by representation, which tackles knowledge relationship mining.