Continuous-Time Link Prediction via Temporal Dependent Graph Neural Network

Qu, Liang; Zhu, Huaisheng; Duan, Qiqi; Shi, Yuhui

doi:10.1145/3366423.3380073

Cited by 56 publications

(27 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One way to simplify the modelling is to convert the dynamic network to an edge-weighted network and then use a static GNN on the edge-weighted network. This is exactly what Temporal Dependent GNN (TDGNN) does [81]. They convert an interaction network to an edge weighted network by using an exponential distribution.…”

Section: B Edge-weighted Modelsmentioning

confidence: 86%

“…GCN [75] or GAT [88]. TDGNN [81] shows a good example of such a scheme by weighting the edges using an exponential distribution, which weights more recent edges higher than old edges.…”

Section: E Discussion and Summarymentioning

confidence: 99%

“…The simplest way to aggregate is to apply an operator, e.g. the inner product [80] (shown in Equation 21), concatenation, mean or Hadamard product [81]. This combines the node embeddings and gives a probability of a link appearing.…”

Section: A Decodersmentioning

confidence: 99%

See 2 more Smart Citations

Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey

2021

View full text Add to dashboard Cite

Dynamic networks are used in a wide range of fields, including social network analysis, recommender systems and epidemiology. Representing complex networks as structures changing over time allow network models to leverage not only structural but also temporal patterns. However, as dynamic network literature stems from diverse fields and makes use of inconsistent terminology, it is challenging to navigate. Meanwhile, graph neural networks (GNNs) have gained a lot of attention in recent years for their ability to perform well on a range of network science tasks, such as link prediction and node classification. Despite the popularity of graph neural networks and the proven benefits of dynamic network models, there has been little focus on graph neural networks for dynamic networks. To address the challenges resulting from the fact that this research crosses diverse fields as well as to survey dynamic graph neural networks, this work is split into two main parts. First, to address the ambiguity of the dynamic network terminology we establish a foundation of dynamic networks with consistent, detailed terminology and notation. Second, we present a comprehensive survey of dynamic graph neural network models using the proposed terminology.INDEX TERMS Dynamic network models, graph neural networks, link prediction, temporal networks.

show abstract

Section: B Edge-weighted Modelsmentioning

confidence: 86%

“…GCN [75] or GAT [88]. TDGNN [81] shows a good example of such a scheme by weighting the edges using an exponential distribution, which weights more recent edges higher than old edges.…”

Section: E Discussion and Summarymentioning

confidence: 99%

See 1 more Smart Citation

Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey

2021

View full text Add to dashboard Cite

show abstract

“…When interactions happen (node or edge adding and removal), models in this class update embeddings of the relevant nodes by aggregating information from their new neighborhoods. In DyGCN ( Cui et al, 2021 ) and TDGNN ( Qu et al, 2020 ), the authors utilize a GCN-based aggregation scheme and propagate changes to higher-order neighbors of interacting nodes. To cope with information asymmetry, the authors of HiLi ( Chen et al, 2021 ) propose to determine the priority of the nodes that receive the latest interaction information.…”

Section: Related Workmentioning

confidence: 99%

Temporal network embedding framework with causal anonymous walks representations

Makarov

Savchenko²,

Korovko

et al. 2022

PeerJ Computer Science

View full text Add to dashboard Cite

Many tasks in graph machine learning, such as link prediction and node classification, are typically solved using representation learning. Each node or edge in the network is encoded via an embedding. Though there exists a lot of network embeddings for static graphs, the task becomes much more complicated when the dynamic (i.e., temporal) network is analyzed. In this paper, we propose a novel approach for dynamic network representation learning based on Temporal Graph Network by using a highly custom message generating function by extracting Causal Anonymous Walks. We provide a benchmark pipeline for the evaluation of temporal network embeddings. This work provides the first comprehensive comparison framework for temporal network representation learning for graph machine learning problems involving node classification and link prediction in every available setting. The proposed model outperforms state-of-the-art baseline models. The work also justifies their difference based on evaluation in various transductive/inductive edge/node classification tasks. In addition, we show the applicability and superior performance of our model in the real-world downstream graph machine learning task provided by one of the top European banks, involving credit scoring based on transaction data.

show abstract

“…From the aspect of recommendation techniques, existing methods are mainly based on collaborative filtering [38,40,47] and deep learning [17,22,59]. Graph Neural Networks (GNN) have recently gained popularity in recommendation systems [5,37,56]. GNNbased recommendation methods [54] typically build the graph with users and items as nodes and interactions as edges, and apply aggregation methods on the graph.…”

Section: Recommendation Systemsmentioning

confidence: 99%

Initialization Matters

Zhang

Liu³

et al. 2021

Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery &Amp; Data Mining

View full text Add to dashboard Cite

Proper initialization is crucial to the optimization and the generalization of neural networks. However, most existing neural recommendation systems initialize the user and item embeddings randomly. In this work, we propose a new initialization scheme for user and item embeddings called Laplacian Eigenmaps with Popularity-based Regularization for Isolated Data (Leporid). Leporid endows the embeddings with information regarding multi-scale neighborhood structures on the data manifold and performs adaptive regularization to compensate for high embedding variance on the tail of the data distribution. Exploiting matrix sparsity, Leporid embeddings can be computed efficiently. We evaluate Leporid in a wide range of neural recommendation models. In contrast to the recent surprising finding that the simple -nearest-neighbor (KNN) method often outperforms neural recommendation systems [8], we show that existing neural systems initialized with Leporid often perform on par or better than KNN. To maximize the effects of the initialization, we propose the Dual-Loss Residual Recommendation (DLR 2 ) network, which, when initialized with Leporid, substantially outperforms both traditional and state-of-the-art neural recommender systems. CCS CONCEPTS• Computing methodologies → Dimensionality reduction and manifold learning; Neural networks; • Information systems → Recommender systems.

show abstract

Continuous-Time Link Prediction via Temporal Dependent Graph Neural Network

Cited by 56 publications

References 16 publications

Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey

Foundations and Modeling of Dynamic Networks Using Dynamic Graph Neural Networks: A Survey

Temporal network embedding framework with causal anonymous walks representations

Initialization Matters

Contact Info

Product

Resources

About