Learning Intents behind Interactions with Knowledge Graph for Recommendation

Wang, Xiang; Huang, Tianyu; Wang, Dingxian; Yuan, Yancheng; Liu, Zhenguang; He, Xiangnan; Chua, Tat-Seng

doi:10.1145/3442381.3450133

Cited by 291 publications

(129 citation statements)

References 34 publications

(73 reference statements)

Supporting

Mentioning

119

Contrasting

Order By: Relevance

“…DR methods [30,37] utilize an imputation model and a prediction model to jointly learn from the MNAR data. Other methods based on information bottleneck [31], meta learning [22], and causal embedding [2,29] have been also explored to address these biases. Among above, IPS and DR has been widely applied to the recommender systems.…”

Section: Counterfactual Learning From Mnar Datamentioning

confidence: 99%

Enhanced Doubly Robust Learning for Debiasing Post-Click Conversion Rate Estimation

Guo

Zou

Liu

et al. 2021

Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

Post-click conversion, as a strong signal indicating the user preference, is salutary for building recommender systems. However, accurately estimating the post-click conversion rate (CVR) is challenging due to the selection bias, i.e., the observed clicked events usually happen on users' preferred items. Currently, most existing methods utilize counterfactual learning to debias recommender systems. Among them, the doubly robust (DR) estimator has achieved competitive performance by combining the error imputation based (EIB) estimator and the inverse propensity score (IPS) estimator in a doubly robust way. However, inaccurate error imputation may result in its higher variance than the IPS estimator. Worse still, existing methods typically use simple model-agnostic methods to estimate the imputation error, which are not sufficient to approximate the dynamically changing model-correlated target (i.e., the gradient direction of the prediction model). To solve these problems, we first derive the bias and variance of the DR estimator. Based on it, a more robust doubly robust (MRDR) estimator has been proposed to further reduce its variance while retaining its double robustness. Moreover, we propose a novel double learning approach for the MRDR estimator, which can convert the error imputation into the general CVR estimation. Besides, we empirically verify that the proposed learning scheme can further eliminate the high variance problem of the imputation learning. To evaluate its effectiveness, extensive experiments are conducted on a semi-synthetic dataset and two real-world datasets. The results demonstrate the superiority of the proposed approach over the state-of-the-art methods. The code is available at https://github.com/guosyjlu/MRDR-DL. CCS CONCEPTS• Information systems → Recommender systems.

show abstract

Section: Counterfactual Learning From Mnar Datamentioning

confidence: 99%

Enhanced Doubly Robust Learning for Debiasing Post-Click Conversion Rate Estimation

Guo

Zou

Liu

et al. 2021

Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

show abstract

“…Disentanglement in Recommendation. Disentangled representation learning in recommendation is largely unexplored until recently [31,42,43,49]. Ma et al [31] propose to learn users' multiple preferences based on Variational Auto-Encoders.…”

Section: Related Workmentioning

confidence: 99%

“…Ma et al [31] propose to learn users' multiple preferences based on Variational Auto-Encoders. Wang et al [42] leverage Knowledge Graph to learn different user intentions and regularize them to be differ from each other. However, most of these works fail to impose specific semantics to the learned multiple representations because of lacking labeled data, i.e.…”

Section: Related Workmentioning

confidence: 99%

Disentangling Long and Short-Term Interests for Recommendation

Zheng,

Gao,

Chang

et al. 2022

Preprint

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

“…To connect the user-POI interaction graph with the knowledge graph, Wang et al [35] proposed that the user-POI interactions are driven by several intents, which can be represented by the distribution of relations in KG. Inspired by it, here we model users' two sets of intentions, geographical and functional intents simultaneously, denoted as I 𝑔 and I 𝑓 .…”

Section: Disentangled Embedding Layermentioning

confidence: 99%

Improving Location Recommendation with Urban Knowledge Graph

Liu¹,

Gao²,

Jin³

et al. 2021

Preprint

View full text Add to dashboard Cite

Location recommendation is defined as to recommend locations (POIs) to users in location-based services. The existing data-driving approaches of location recommendation suffer from the limitation of the implicit modeling of the geographical factor, which may lead to sub-optimal recommendation results. In this work, we address this problem by introducing knowledge-driven solutions. Specifically, we first construct the Urban Knowledge Graph (UrbanKG) with geographical information and functional information of POIs. On the other side, there exist a fact that the geographical factor not only characterizes POIs but also affects user-POI interactions.To address it, we propose a novel method named UKGC. We first conduct information propagation on two sub-graphs to learn the representations of POIs and users. We then fuse two parts of representations by counterfactual learning for the final prediction. Extensive experiments on two real-world datasets verify that our method can outperform the state-of-the-art methods.

show abstract

Learning Intents behind Interactions with Knowledge Graph for Recommendation

Cited by 291 publications

References 34 publications

Enhanced Doubly Robust Learning for Debiasing Post-Click Conversion Rate Estimation

Enhanced Doubly Robust Learning for Debiasing Post-Click Conversion Rate Estimation

Disentangling Long and Short-Term Interests for Recommendation

Improving Location Recommendation with Urban Knowledge Graph

Contact Info

Product

Resources

About