Scalable Temporal Latent Space Inference for Link Prediction in Dynamic Social Networks

By Linhong Zhu, Dong Guo, Junming Yin, Greg Ver Steeg, Aram Galstyan. 2014

210 citations recorded by the source corpus.

We propose a temporal latent space model for link prediction in dynamic social networks, where the goal is to predict links over time based on a sequence of previous graph snapshots. The model assumes that each user lies in an unobserved latent space, and interactions are more likely to occur between similar users in the latent space representation. In addition, the model allows each user to gradually move its position in the latent space as the network structure evolves over time. We present a global optimization algorithm to effectively infer the temporal latent space. Two alternative optimization algorithms with local and incremental updates are also proposed, allowing the model to scale to larger networks without compromising prediction accuracy. Empirically, we demonstrate that our model, when evaluated on a number of real-world dynamic networks, significantly outperforms existing approaches for temporal link prediction in terms of both scalability and predictive power.

Subfields: Statistical and Nonlinear Physics, Transportation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistical and Nonlinear Physics

In Transportation

Citations in the corpus

Recurrent Event Network for Reasoning over Temporal Knowledge Graphs (2019, background): “Furthermore, extending existing dynamic network embedding models [8, 29, 16, 8] to perform TKG reasoning is non-trivial—they do not capture multi-relational entity interaction patterns and require expensive re-training when doing multi-step prediction over future time stamps.”
Relational Representation Learning for Dynamic (Knowledge) Graphs: A Survey (2019, methodology): “A common approach to leverage the temporal aspect of DTDGs is to use time as a regularizer to impose a smoothness constraint over the embeddings of each node over consecutive snapshots [31, 43, 120, 86, 242, 259, 257].”
dyngraph2vec: Capturing Network Dynamics using Dynamic Graph Representation Learning (2018, methodology): “We use a weighting matrix B to weight the reconstruction of observed edges higher than unobserved links as traditionally used in the literature (Wang, Cui, and Zhu 2016).”
Dynamic Network Embedding by Modeling Triadic Closure Process (2018, background): “We try different values of TNE’s parameter λ (option “-l” in the implementation) and report the best performance.”
DySAT: Deep Neural Representation Learning on Dynamic Graphs via Self-Attention Networks (2020, methodology): “Temporal smoothness methods ensure embedding stability across consecutive time-steps [27, 40].”