Node Embedding over Temporal Graphs

Abstract

In this work, we present a method for node embedding in temporal graphs. We propose an algorithm that learns the evolution of a temporal graph's nodes and edges over time and incorporates this dynamics in a temporal node embedding framework for different graph prediction tasks. We present a joint loss function that creates a temporal embedding of a node by learning to combine its historical temporal embeddings, such that it optimizes per given task (e.g., link prediction). The algorithm is initialized using static node embeddings, which are then aligned over the representations of a node at different time points, and eventually adapted for the given task in a joint optimization. We evaluate the effectiveness of our approach over a variety of temporal graphs for the two fundamental tasks of temporal link prediction and multi-label node classification, comparing to competitive baselines and algorithmic alternatives. Our algorithm shows performance improvements across many of the datasets and baselines and is found particularly effective for graphs that are less cohesive, with a lower clustering coefficient.

Susceptible-infected-spreading-based network embedding in static and temporal networks

“…CTDNE [31] is a temporal network embedding algorithm based on random walks. We also choose tNodeEmbed [41] as a baseline model, which uses a different learning model from the above models. It learns temporal network embedding by combining static node embeddings learned from Node2Vec with Recurrent Neural Networks.…”

Temporal Latent Space Modeling for Community Prediction

“…This can be employed for different underlying applications such as link prediction and node classification [13]. Various works such as Temporal tensorfactorization [11,34], and neural embeddings [14,27,35] have been proposed in this respect. For instance, Singer et al [27] extend the prior neural-based embedding approaches on static graphs, e.g., node2vec [15], to temporal graphs.…”

“…Various works such as Temporal tensorfactorization [11,34], and neural embeddings [14,27,35] have been proposed in this respect. For instance, Singer et al [27] extend the prior neural-based embedding approaches on static graphs, e.g., node2vec [15], to temporal graphs. They propose a semi-supervised algorithm, namely tNodeEmbed, that learns to combine a node's historical temporal embeddings into a final embedding such that it can optimize for a given underlying task, e.g., link prediction.…”

“…First, contrary to non-temporal methods [31][32][33], our approach incorporates temporality. Second, in contrast to Zhu et al's method [34] and the likes [11,14,27,35] that focus on dynamics of social network structure, our approach employs dynamics of social content. Third, although we use temporal information to predict future users' topical interests similar to regression methods, we train only one model for all users and, thus, significantly reduce computational cost compared to regression techniques.…”

weg2vec: Event embedding for temporal networks

“…At the base of many methods there is the modification of the standard representation of the temporal network, whether it is in the form of a list of events, a tensor [9] or a supra-adjacency matrix [30]. All of these methods [31,32,33] commonly aim to solve a node embedding problem by locally sampling the temporal-structural neighbourhood of nodes to create contexts, which they feed to a Skip-Gram learning architecture borrowed from the text representation literature [34]. As a solution, they build a sequence of correlated/updated embeddings of network snapshots, which consider short term history of the network backward in time.…”