Neighbour Interaction based Click-Through Rate Prediction via Graph-masked TransformerCitation formats

  • External authors:
  • Erxue Min
  • Yu Rong
  • Tingyang Xu
  • Yatao Bian
  • Da Luo
  • Kangyi Lin
  • Junzhou Huang
  • Peilin Zhao

Standard

Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer. / Min, Erxue; Rong, Yu ; Xu, Tingyang et al.

2022. 353-362 Paper presented at SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval.

Research output: Contribution to conferencePaperpeer-review

Harvard

Min, E, Rong, Y, Xu, T, Bian, Y, Luo, D, Lin, K, Huang, J, Ananiadou, S & Zhao, P 2022, 'Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer', Paper presented at SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, 11/07/22 - 15/07/22 pp. 353-362. https://doi.org/10.1145/3477495.3532031

APA

Min, E., Rong, Y., Xu, T., Bian, Y., Luo, D., Lin, K., Huang, J., Ananiadou, S., & Zhao, P. (2022). Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer. 353-362. Paper presented at SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. https://doi.org/10.1145/3477495.3532031

Vancouver

Min E, Rong Y, Xu T, Bian Y, Luo D, Lin K et al.. Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer. 2022. Paper presented at SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. https://doi.org/10.1145/3477495.3532031

Author

Min, Erxue ; Rong, Yu ; Xu, Tingyang et al. / Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer. Paper presented at SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval.10 p.

Bibtex

@conference{1c8b7d0d4ae84de1ba17160da9304455,
title = "Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer",
abstract = "Click-Through Rate (CTR) prediction, which aims to estimate the probability that a user will click an item, is an essential component of online advertising. Existing methods mainly attempt to mine user interests from users' historical behaviours, which contain users' directly interacted items. Although these methods have made great progress, they are often limited by the recommender system's direct exposure and inactive interactions, and thus fail to mine all potential user interests. To tackle these problems, we propose Neighbor-Interaction based CTR prediction (NI-CTR), which considers this task under a Heterogeneous Information Network (HIN) setting. In short, Neighbor-Interaction based CTR prediction involves the local neighborhood of the target user-item pair in the HIN to predict their linkage. In order to guide the representation learning of the local neighbourhood, we further consider different kinds of interactions among the local neighborhood nodes from both explicit and implicit perspective, and propose a novel Graph-Masked Transformer (GMT) to effectively incorporates these kinds of interactions to produce highly representative embeddings for the target user-item pair. Moreover, in order to improve model robustness against neighbour sampling, we enforce a consistency regularization loss over the neighbourhood embedding. We conduct extensive experiments on two real-world datasets with millions of instances and the experimental results show that our proposed method outperforms state-of-the-art CTR models significantly. Meanwhile, the comprehensive ablation studies verify the effectiveness of every component of our model. Furthermore, we have deployed this framework on the WeChat Official Account Platform with billions of users. The online A/B tests demonstrate an average CTR improvement of 21.9% against all online baselines.",
author = "Erxue Min and Yu Rong and Tingyang Xu and Yatao Bian and Da Luo and Kangyi Lin and Junzhou Huang and Sophia Ananiadou and Peilin Zhao",
year = "2022",
month = jul,
day = "7",
doi = "https://doi.org/10.1145/3477495.3532031",
language = "English",
pages = "353--362",
note = "SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR'22 ; Conference date: 11-07-2022 Through 15-07-2022",
url = "https://dl.acm.org/doi/abs/10.1145/3477495.3532031",

}

RIS

TY - CONF

T1 - Neighbour Interaction based Click-Through Rate Prediction via Graph-masked Transformer

AU - Min, Erxue

AU - Rong, Yu

AU - Xu, Tingyang

AU - Bian, Yatao

AU - Luo, Da

AU - Lin, Kangyi

AU - Huang, Junzhou

AU - Ananiadou, Sophia

AU - Zhao, Peilin

N1 - Conference code: 45

PY - 2022/7/7

Y1 - 2022/7/7

N2 - Click-Through Rate (CTR) prediction, which aims to estimate the probability that a user will click an item, is an essential component of online advertising. Existing methods mainly attempt to mine user interests from users' historical behaviours, which contain users' directly interacted items. Although these methods have made great progress, they are often limited by the recommender system's direct exposure and inactive interactions, and thus fail to mine all potential user interests. To tackle these problems, we propose Neighbor-Interaction based CTR prediction (NI-CTR), which considers this task under a Heterogeneous Information Network (HIN) setting. In short, Neighbor-Interaction based CTR prediction involves the local neighborhood of the target user-item pair in the HIN to predict their linkage. In order to guide the representation learning of the local neighbourhood, we further consider different kinds of interactions among the local neighborhood nodes from both explicit and implicit perspective, and propose a novel Graph-Masked Transformer (GMT) to effectively incorporates these kinds of interactions to produce highly representative embeddings for the target user-item pair. Moreover, in order to improve model robustness against neighbour sampling, we enforce a consistency regularization loss over the neighbourhood embedding. We conduct extensive experiments on two real-world datasets with millions of instances and the experimental results show that our proposed method outperforms state-of-the-art CTR models significantly. Meanwhile, the comprehensive ablation studies verify the effectiveness of every component of our model. Furthermore, we have deployed this framework on the WeChat Official Account Platform with billions of users. The online A/B tests demonstrate an average CTR improvement of 21.9% against all online baselines.

AB - Click-Through Rate (CTR) prediction, which aims to estimate the probability that a user will click an item, is an essential component of online advertising. Existing methods mainly attempt to mine user interests from users' historical behaviours, which contain users' directly interacted items. Although these methods have made great progress, they are often limited by the recommender system's direct exposure and inactive interactions, and thus fail to mine all potential user interests. To tackle these problems, we propose Neighbor-Interaction based CTR prediction (NI-CTR), which considers this task under a Heterogeneous Information Network (HIN) setting. In short, Neighbor-Interaction based CTR prediction involves the local neighborhood of the target user-item pair in the HIN to predict their linkage. In order to guide the representation learning of the local neighbourhood, we further consider different kinds of interactions among the local neighborhood nodes from both explicit and implicit perspective, and propose a novel Graph-Masked Transformer (GMT) to effectively incorporates these kinds of interactions to produce highly representative embeddings for the target user-item pair. Moreover, in order to improve model robustness against neighbour sampling, we enforce a consistency regularization loss over the neighbourhood embedding. We conduct extensive experiments on two real-world datasets with millions of instances and the experimental results show that our proposed method outperforms state-of-the-art CTR models significantly. Meanwhile, the comprehensive ablation studies verify the effectiveness of every component of our model. Furthermore, we have deployed this framework on the WeChat Official Account Platform with billions of users. The online A/B tests demonstrate an average CTR improvement of 21.9% against all online baselines.

U2 - https://doi.org/10.1145/3477495.3532031

DO - https://doi.org/10.1145/3477495.3532031

M3 - Paper

SP - 353

EP - 362

T2 - SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval

Y2 - 11 July 2022 through 15 July 2022

ER -