Mohammad Mehdi Keikha - PhD Candidate, University of Tehran, Tehran, Iran Faculty member, University of Sistan and Baluchestan, Zahedan, Iran
Maseud Rahgozar - Associate Professor, University of Tehran, Tehran

Link prediction on social networks is one of the issues that has attracted many researchers in recent years. In this problem, missing and future links are predicted by using existing links in the. One of the newest approaches to this problem is the use of deep learning to extract the vector of the features of each node and then find missing and future links. This paper presents a method for learning the vector representation of network nodes based on the information of the nodes adjacent to each node in the social network and the various links present on the network. The results show that the proposed method provides good results for link prediction in comparison with other methods.

Link prediction, Deep Learning, feature vector of node, network communities, social networks