AFTLP: A Comparative Analytical Framework for Temporal Link Prediction

Link prediction is an important task in many application domains. Widespread researches have been done to extract information from network's entities and relations. Temporal Link prediction is a task of predicting unseen links that will be formed in the future based on different snapshots of the network. This paper proposes a framework in which the concerns and challenges of temporal link prediction are discussed. All major and recent methodologies in temporal link prediction are classified. This presented classification is analyzed based on the discussed challenges. Accordingly, this framework provides an appropriate platform for comparing and analyzing different approaches and challenges, that will encourage future research in developing more powerful methods.