Abstract:Link prediction on dynamic networks has been extensively studied and widely applied in various applications. However, existing methods only consider either the network structure or the temporal information, ignoring the potentialities of using both types of information together to comprehend the complex behaviors of dynamic networks. Moreover, temporal unlink prediction, which also plays an important role in the evolution of social networks, has not been paid much attention. Accurately predicting the links and unlinks on the future network greatly contributes to the network analysis that uncovers more latent relations between nodes. In this work, we assume that there are two kinds of relations between nodes, namely, long-term relations and short-term relations, and we propose an effective algorithm called LULS for temporal link prediction and unlink prediction based on such relations. Specifically, for each snapshot of a dynamic network, LULS first collects higher order structures as two topological matrices by applying short random walks. Then, LULS initializes and optimizes a global matrix and a sequence of temporary matrices for all the snapshots by using nonnegative matrix factorization (NMF) based on the topological matrices, where the global matrix denotes long-term relations and the temporary matrices represent short-term relations of snapshots. Finally, LULS calculates the similarity matrix of the future snapshot and predicts the links and unlinks for the future network. In addition, we further improve the prediction results by using graph regularization constraints to enhance the global matrix, resulting in that the global matrix contains a wealth of topological information and temporal information. The conducted experiments on real-world networks illustrate that LULS outperforms other baselines for both link prediction and unlink prediction tasks.

Link Prediction in Dynamic Networks Based on Machine Learning

A Supervised Learning Approach To Link Prediction In Dynamic Networks

Link Prediction Via Supervised Dynamic Network Formation.

Link Prediction in Dynamic Networks Based on the Attraction Force Between Nodes

Link Prediction Based on Dynamic Weighted Social Attribute Network.

Dynamic link prediction by learning the representation of node-pair via graph neural networks

A Deep Learning Approach to Link Prediction in Dynamic Networks

Link prediction of time-evolving network based on node ranking

Link Prediction in Dynamic Social Networks Combining Entropy, Causality, and a Graph Convolutional Network Model

Link Prediction with Spatial and Temporal Consistency in Dynamic Networks.

Evolving Networks-Using Past Structure to Predict the Future

A Dynamic Logistic Regression for Network Link Prediction

Dynamic Link Prediction by Integrating Node Vector Evolution and Local Neighborhood Representation

Effective Model Integration Algorithm for Improving Link and Sign Prediction in Complex Networks

Predicting the Evolution of Complex Networks Via Similarity Dynamics

Link Prediction and Unlink Prediction on Dynamic Networks

A Universal Method Based on Structure Subgraph Feature for Link Prediction over Dynamic Networks

Dynamic Network Link Prediction Based on Random Walking and Time Aggregation

Deep Dynamic Network Embedding for Link Prediction

Link Prediction Based on Linear Dynamical Response

Link Prediction for Existing Links in Dynamic Networks Based on the Attraction Force