Sequential Recommendation with Graph Neural Networks

Chang, Jianxin; Gao, Chen; Zheng, Yu; Hui, Yiqun; Niu, Yifan; Song, Yang; Jin, Depeng; Li, Yong

doi:10.1145/3404835.3462968

Cited by 266 publications

(75 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The historical sessions are completely different from the dialogue/conversation history used in previous CRS works [2,39,41], as their "historical" information is actually the historical turns of the current dialogue session. The usages of historical sessions in CRS are also largely different from the historical user-item interactions [1,31] in traditional recommendations, since the recommendation in CRS is strongly constrained by the current user intentions, while traditional recommendations are not. Thus, the main goal of historical sessions learner is: learning current-related and regular information from historical sessions, without impeding the current session information.…”

Section: Motivations and Discussion On Uccrmentioning

confidence: 99%

“…Specifically, UCCR learns the multi-aspect user preferences mainly from three information sources, including the user's current dialogue session, historical dialogues sessions, and look-alike users. UCCR mainly consists of four parts: (1) We first design a historical session learner to capture users' diverse preferences in their historical sessions besides learning from the current session. Precisely, we extract multi-view user preferences from the dialogues, including the word-level semantic view, entity-level knowledge view, and item-level consuming view.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

User-Centric Conversational Recommendation with Multi-Aspect User Modeling

Li,

Xie,

Zhu

et al. 2022

Preprint

View full text Add to dashboard Cite

Conversational recommender systems (CRS) aim to provide highquality recommendations in conversations. However, most conventional CRS models mainly focus on the dialogue understanding of the current session, ignoring other rich multi-aspect information of the central subjects (i.e., users) in recommendation. In this work, we highlight that the user's historical dialogue sessions and look-alike users are essential sources of user preferences besides the current dialogue session in CRS. To systematically model the multi-aspect information, we propose a User-Centric Conversational Recommendation (UCCR) model, which returns to the essence of user preference learning in CRS tasks. Specifically, we propose a historical session learner to capture users' multi-view preferences from knowledge, semantic, and consuming views as supplements to the current preference signals. A multi-view preference mapper is conducted to learn the intrinsic correlations among different views in current and historical sessions via self-supervised objectives. We also design a temporal look-alike user selector to understand users via their similar users. The learned multi-aspect multi-view user preferences are then used for the recommendation and dialogue generation. In experiments, we conduct comprehensive evaluations on both Chinese and English CRS datasets. The significant improvements over competitive models in both recommendation and dialogue generation verify the superiority of UCCR. The source code will be available on https://github.com/lisk123/UCCR.

show abstract

Section: Motivations and Discussion On Uccrmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

User-Centric Conversational Recommendation with Multi-Aspect User Modeling

Li,

Xie,

Zhu

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…With respect to long-term interests modeling, we include NCF [16], DIN [51] and LightGCN [14]. For short-term interests modeling, we compare with Caser [41], GRU4REC [17], DIEN [50], SASRec [20] and SURGE [6]. We also include SLi-Rec [47] which is the state-of-the-art model of LS-term interests modeling.…”

Section: Methodsmentioning

confidence: 99%

“…We use two datasets to conduct experiments, including a public e-commerce dataset and an industrial short-video dataset, which are also adopted by the SOTA sequential recommendation model, SURGE [6]. Both of them are in million scale and collected from real-world applications.…”

Section: A Appendix A1 Datasetsmentioning

confidence: 99%

Disentangling Long and Short-Term Interests for Recommendation

Zheng,

Gao,

Chang

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Modeling user's long-term and short-term interests is crucial for accurate recommendation. However, since there is no manually annotated label for user interests, existing approaches always follow the paradigm of entangling these two aspects, which may lead to inferior recommendation accuracy and interpretability. In this paper, to address it, we propose a Contrastive learning framework to disentangle Long and Short-term interests for Recommendation (CLSR) with self-supervision. Specifically, we first propose two separate encoders to independently capture user interests of different time scales. We then extract long-term and short-term interests proxies from the interaction sequences, which serve as pseudo labels for user interests. Then pairwise contrastive tasks are designed to supervise the similarity between interest representations and their corresponding interest proxies. Finally, since the importance of long-term and short-term interests is dynamically changing, we propose to adaptively aggregate them through an attentionbased network for prediction. We conduct experiments on two large-scale real-world datasets for e-commerce and short-video recommendation. Empirical results show that our CLSR consistently outperforms all state-of-the-art models with significant improvements: GAUC is improved by over 0.01, and NDCG is improved by over 4%. Further counterfactual evaluations demonstrate that stronger disentanglement of long and short-term interests is successfully achieved by CLSR. The code and data are available at https://github.com/tsinghua-fib-lab/CLSR.

show abstract

“…The core idea is that, instead of using a predefined distance function, we leverage metric learning [74] to learn a distance metric for the input space of data (i.e., the node embedding matrix H) from the adjacency matrix A that preserves the node relationships (i.e., A is used to supervise the distance learning). In this paper, we adopt a weighted cosine distance (defined in Equation 2) [8,83] as our learnable distance function φ.…”

Section: Graph Metric Learningmentioning

confidence: 99%

Finding MNEMON: Reviving Memories of Node Embeddings

Shen¹,

Han²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Previous security research efforts orbiting around graphs have been exclusively focusing on either (de-)anonymizing the graphs or understanding the security and privacy issues of graph neural networks. Little attention has been paid to understand the privacy risks of integrating the output from graph embedding models (e.g., node embeddings) with complex downstream machine learning pipelines. In this paper, we fill this gap and propose a novel model-agnostic graph recovery attack that exploits the implicit graph structural information preserved in the embeddings of graph nodes. We show that an adversary can recover edges with decent accuracy by only gaining access to the node embedding matrix of the original graph without interactions with the node embedding models. We demonstrate the effectiveness and applicability of our graph recovery attack through extensive experiments.

show abstract

Sequential Recommendation with Graph Neural Networks

Cited by 266 publications

References 30 publications

User-Centric Conversational Recommendation with Multi-Aspect User Modeling

User-Centric Conversational Recommendation with Multi-Aspect User Modeling

Disentangling Long and Short-Term Interests for Recommendation

Finding MNEMON: Reviving Memories of Node Embeddings

Contact Info

Product

Resources

About