Temporal Heterogeneous Interaction Graph Embedding for Next-Item Recommendation

Ji, Yugang; Yin, Mingyang; Yang, Hongxia; Wang, Xiangwei; Jia, Tianrui; Shi, Chuan

doi:10.1007/978-3-030-67664-3_19

Cited by 15 publications

(7 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To capture structural information of user-item interactions as much as possible, researcher employed hyper-graph [40]- [43], tripartite graph [44], and heterogeneous interaction graphs [45]. However, they are not efficient for training as bipartite graph [20], which is our choice.…”

Section: Sequential Recommendation (Abbrev Sr)mentioning

confidence: 99%

DraG4Rec: A Symmetrical View of User Preference and Item Popularity

Gao¹,

Hou²,

Ying³

2023

Preprint

View full text Add to dashboard Cite

<p>Utilizing user-item interaction dynamics is crucial for providing efficient and precise sequential recommendations (SR). Besides earlier works based on Matrix factorization, Markov Chains, and RNNs, GCNs taking advantage of attention mechanism and Knowledge-enhanced NNs digging heterogenous auxiliary information recently made improvements. However, there still exist some problems: 1) existing works mainly focus on capturing user preference without considering the importance of item popularity; 2) introducing too much auxiliary information denies the independence of interaction behavior; 3) favoring high-order chronological history with heavy memory while ranking score referring different bias is unappreciated. To tackle these problems, we proposed DraG4Rec (Dual rating enhanced attention-based GCN for Recommendation), which makes symmetry of user preference and item popularity manifest great value. Specifically, we first merge ranking score, time, and position encoding for user/item representation in a bipartite graph. Then, an edge-view message-passing mechanism is imported into the attention-based GCN to use implicit 2-order historical information with less memory. Finally, user preference and item popularity are learned jointly. Extensive experiments on dense and sparse real-world datasets demonstrate the superior performance of our framework over state-of-the-art baselines regarding two commonly-used metrics: i.e., Recall and NDCG. The ablation study illustrates the boosting effect of different components. </p>

show abstract

Section: Sequential Recommendation (Abbrev Sr)mentioning

confidence: 99%

DraG4Rec: A Symmetrical View of User Preference and Item Popularity

Gao¹,

Hou²,

Ying³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…[35] proposes DyHATR, which uses the hierarchical attention model to capture the heterogeneity and introduces the temporal attentive GRU/LSTM to model the evolutionary patterns among snapshots. While [36] focus on next-item recommendation problem, they learn embedings on temporal heterogenous User-Item bipartite network. This kind of network has multiple types of edges, representing different interaction behaviors.…”

Section: Rnn Based Modelmentioning

confidence: 99%

“…[24] proposes BurstGraph to capture unexpected bursty changes and [32] uses a coupled RNN to update embeddings for users and items. [69] completes social recommendation based on session while [36] completes next-item recommendation on temporal heterogeneous net-work.…”

Section: • Anomaly Detectionmentioning

confidence: 99%

“…Heterogenous network has shown great potential to provide richer information with various node types and edge types. The static embedding on heterogenous network is well studied recently [51,75,76] while the research on heterogeneous dynamic graphs is just getting started [26,36,35,77]. How to further explore the relationship between various types of nodes and edges in the process of network change, and to mine the unique time patterns of heterogeneous graphs is a direction worth studying.…”

Section: Conclusion and Future Directionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Network Embedding Survey

Xue¹,

Zhong²,

Li³

et al. 2021

Preprint

View full text Add to dashboard Cite

Since many real world networks are evolving over time, such as social networks and user-item networks, there are increasing research efforts on dynamic network embedding in recent years. They learn node representations from a sequence of evolving graphs but not only the latest network, for preserving both structural and temporal information from the dynamic networks. Due to the lack of comprehensive investigation of them, we give a survey of dynamic network embedding in this paper. Our survey inspects the data model, representation learning technique, evaluation and application of current related works and derives common patterns from them. Specifically, we present two basic data models, namely, discrete model and continuous model for dynamic networks. Correspondingly, we summarize two major categories of dynamic network embedding techniques, namely, structural-first and temporal-first that are adopted by most related works. Then we build a taxonomy that refines the category hierarchy by typical learning models. The popular experimental data sets and applications are also summarized. Lastly, we have a discussion of several distinct research topics in dynamic network embedding.

show abstract

“…Many existing works have been proposed to tackle the temporal link prediction problem [1,2,5,7,8,10,14,[16][17][18][19][20]. TGN [6,9], as a representative work, presents a general deep learning framework for this task based on memory mechanisms and graph-based operators.…”

Section: Introductionmentioning

confidence: 99%

HTGN-BTW: Heterogeneous Temporal Graph Network with Bi-Time-Window Training Strategy for Temporal Link Prediction

Yue¹,

Du²,

Fu³

et al. 2022

Preprint

View full text Add to dashboard Cite

With the development of temporal networks such as E-commerce networks and social networks, the issue of temporal link prediction has attracted increasing attention in recent years. The Temporal Link Prediction task of WSDM Cup 2022 1 expects a single model that can work well on two kinds of temporal graphs simultaneously, which have quite different characteristics and data properties, to predict whether a link of a given type will occur between two given nodes within a given time span. Our team, named as nothing here, regards this task as a link prediction task in heterogeneous temporal networks and proposes a generic model, i.e., Heterogeneous Temporal Graph Network (HTGN), to solve such temporal link prediction task with the unfixed time intervals and the diverse link types. That is, HTGN can adapt to the heterogeneity of links and the prediction with unfixed time intervals within an arbitrary given time period. To train the model, we design Bi-Time-Window training strategy (BTW) which has two kinds of mini-batches from two kinds of time windows. As a result, for the final test, we achieved an AUC of 0.662482 on dataset A, an AUC of 0.906923 on dataset B, and won 2nd place with an Average T-scores of 0.628942.

show abstract

Temporal Heterogeneous Interaction Graph Embedding for Next-Item Recommendation

Cited by 15 publications

References 20 publications

DraG4Rec: A Symmetrical View of User Preference and Item Popularity

DraG4Rec: A Symmetrical View of User Preference and Item Popularity

Dynamic Network Embedding Survey

HTGN-BTW: Heterogeneous Temporal Graph Network with Bi-Time-Window Training Strategy for Temporal Link Prediction

Contact Info

Product

Resources

About