A Survey of Link Prediction in Information Networks

Most information networks containing more and more information consist of a large number of interacting objects, while more and more researchers pay their attentions to information mining in networks. As an important topic of information mining in networks, link prediction is intended to predict the possibility of future links between two objects in the network by analyzing existing network information. Since the Internet of Things (IoT) is a special information network, research on link prediction in information networks has important implications for the topology and extension of IoT. In this paper, we provide a survey of link prediction in information networks. We will introduce basic concepts and definitions of information networks, summarize the latest link prediction methods in information networks, including link prediction methods in single layer networks, link prediction methods in multiplex networks, and link prediction methods in heterogeneous networks, point out the existing problems and the future research direction.