Abstract
Link prediction aims to predict the missing edge or the edge that may be generated in the future. The key to link prediction is to obtain the characteristic information with strong representation for nodes. In this work, aiming at the problem that the existing link prediction methods fail to take full account of the high-order connection mode of nodes, we present a new representation learning-based approach called OC-GAE (Orbit Counting and Graph Auto-Encoder) that considers rich subgraph structure around nodes. Firstly, the number of orbits on subgraphs is calculated as the high-order structural features of the nodes; then, the number of orbits is used as the input of Graph Auto-Encoder to learn the efficient representation of the nodes; finally, the network adjacency matrix is reconstructed by the learned representation to realize the link prediction. By comparing with five classical link prediction methods and two mainstream network representation learning methods on four real network datasets, the effectiveness of the proposed method and the prediction accuracy are proved to be optimal in general.
Highlights
Inferring missing links or predicting future ones based on the currently observed network is known as the link prediction problem, which has tremendous real-world applications, such as predicting user relationships in social networks [1], recommending items to users in recommendation systems [2, 3], predicting drug-target interactions [4], etc.The most intuitive assumption of link prediction is that the more similar the two nodes are, the more likely they are to have edges
The link prediction methods based on node attributes mainly use the external attributes and label information of nodes to describe the similarity between nodes, while the link prediction methods based on network structure measure the structural similarity of nodes based on the network topology
This paper proposes a link prediction method named OC-GAE (Orbit Counting and Graph Auto-Encoder) based on orbits and GAE with the help of the latest research on graphlet and graph representation learning
Summary
Inferring missing links or predicting future ones based on the currently observed network is known as the link prediction problem, which has tremendous real-world applications, such as predicting user relationships in social networks [1], recommending items to users in recommendation systems [2, 3], predicting drug-target interactions [4], etc. It can be seen that the local link prediction method based on common neighbors cannot fully represent the similarity between nodes in some cases, still need more fine-grained topology information. The problems to be solved for link prediction through the high-level structure of the network are: 1 how to mine the graphlet structure of nodes to capture higher-order patterns; 2 how to design a model to learn the characteristics of graphlet structure To address these problems, this paper proposes a link prediction method named OC-GAE (Orbit Counting and Graph Auto-Encoder) based on orbits and GAE with the help of the latest research on graphlet and graph representation learning. The rest of this paper is organized as follows: Section 2 introduces the related works; Section 3 introduces the link prediction method based on the orbit counting and GAE; Section 4 shows the experimental results; Section 5 concludes this article and points out the future works
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
