Article Code : 13960203174625750

Article Title : A content-based link prediction algorithm in scientific social networks

Keywords :

Journal Number : 19 Summer 2017

Visited : 298

Files : 397 KB

List of Authors

  Full Name Email Grade Degree Corresponding Author
1 hosna solaimannezhad Graduate M.Sc
2 omid fatemi Assistant Professor PhD


Prediction of the collaboration of two authors by their research interests is one of the issues to enhance group researches. One of the main topics in the analysis of social networks is link prediction. One type of social networks as co-authors social network is one of the highly applied datasets. If we denote a social network by a graph, link prediction means the prediction of the edges between the nodes of network in future. The output of link prediction algorithms is used in different fields as recommender systems. There are a few studies on link prediction using the content issued by nodes to predict the link. In this study, a new link prediction algorithm is developed based on the interest of people. By extracting the working fields of the authors via the analysis of published papers by them, this algorithm predicts their ties in future. The results of tests on SID dataset as co-author dataset show that the presented algorithm outperforms all the structure-based link prediction algorithms. Finally, the reasons of efficiency of algorithm are analyzed and presented.