Learning to Retrieve User Behaviors for Click-through Rate Estimation

被引:6
|
作者
Qin, Jiarui [1 ]
Zhang, Weinan [1 ]
Su, Rong [2 ]
Liu, Zhirong [2 ]
Liu, Weiwen [2 ]
Zhao, Guangpeng [3 ]
Li, Hao [3 ]
Tang, Ruiming [2 ]
He, Xiuqiang [2 ]
Yu, Yong [1 ]
机构
[1] Shanghai Jiao Tong Univ, Shanghai 200240, Peoples R China
[2] Huawei Noahs Ark Lab, Shenzhen 518129, Peoples R China
[3] Huawei Consumer Cloud Serv Dept, Shenzhen 518129, Peoples R China
关键词
CTR estimation; information retrieval; sequential user behavior modeling;
D O I
10.1145/3579354
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Click-through rate (CTR) estimation plays a crucial role in modern online personalization services. It is essential to capture users' drifting interests by modeling sequential user behaviors to build an accurate CTR estimation model. However, as the users accumulate a large amount of behavioral data on the online platforms, the current CTR models have to truncate user behavior sequences and utilize the most recent behaviors, which leads to a problem that sequential patterns such as periodicity or long-term dependency are not contained in the recent behaviors but in far back history. However, it is non-trivial to model the entire user sequence by directly using it for two reasons. Firstly, the very long input sequences will make online inference time and system load infeasible. Secondly, the very long sequences contain much noise, thus making it difficult for CTR models to capture useful patterns effectively. To tackle this issue, we consider it from the input data perspective instead of designing more sophisticated yet complex models. As the entire user behavior sequence contains much noise, it is unnecessary to input the entire sequence. Instead, we could just retrieve only a small part of it as the input to the CTR model. In this article, we propose the User Behavior Retrieval (UBR) framework which aims at learning to retrieve the most informative user behaviors according to each CTR estimation request. Retrieving only a small set of behaviors could alleviate the two problems of utilizing very long sequences (i.e., inference efficiency and noisy input). The distinguishing property of UBR is that it supports arbitrary and learnable retrieval functions instead of utilizing a fixed pre-defined function, which is different from the current retrieval-based methods. Offline evaluations on three large-scale real-world datasets demonstrate the superiority and efficacy of the UBR framework. We further deploy UBR at the Huawei App Store, where it achieves 6.6% of eCPM gain in the online A/B test and now serves the main traffic in the Huawei App Store advertising scenario.
引用
收藏
页数:31
相关论文
共 50 条
  • [31] Ad Click-Through Rate Prediction: A Survey
    Gu, Liqiong
    DATABASE SYSTEMS FOR ADVANCED APPLICATIONS: DASFAA 2021 INTERNATIONAL WORKSHOPS, 2021, 12680 : 140 - 153
  • [32] Click fraud resistant methods for learning click-through rates
    Immorlica, N
    Jain, K
    Mahdian, M
    Talwar, K
    INTERNET AND NETWORK ECONOMICS, PROCEEDINGS, 2005, 3828 : 34 - 45
  • [33] XCrossNet: Feature Structure-Oriented Learning for Click-Through Rate Prediction
    Yu, Runlong
    Ye, Yuyang
    Liu, Qi
    Wang, Zihan
    Yang, Chunfeng
    Hu, Yucheng
    Chen, Enhong
    ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PAKDD 2021, PT II, 2021, 12713 : 436 - 447
  • [34] PeNet : A feature excitation learning approach to advertisement click-through rate prediction
    Yin, Yunfei
    Ochieng, Nyambega David
    Sun, Jingqin
    Bao, Xianjian
    Wang, Zhuowei
    NEURAL NETWORKS, 2024, 172
  • [35] Improving Click-Through Rate Prediction Accuracy in Online Advertising by Transfer Learning
    Su, Yuhan
    Jin, Zhongming
    Chen, Ying
    Sun, Xinghai
    Yang, Yaming
    Qiao, Fangzheng
    Xia, Fen
    Xu, Wei
    2017 IEEE/WIC/ACM INTERNATIONAL CONFERENCE ON WEB INTELLIGENCE (WI 2017), 2017, : 1018 - 1025
  • [36] A Hierarchical Extreme Learning Machine Algorithm for Advertisement Click-Through Rate Prediction
    Zhang, Sen
    Liu, Zheng
    Xiao, Wendong
    IEEE ACCESS, 2018, 6 : 50641 - 50647
  • [37] RLNF: Reinforcement Learning based Noise Filtering for Click-Through Rate Prediction
    Zhao, Pu
    Luo, Chuan
    Zhou, Cheng
    Qiao, Bo
    He, Jiale
    Zhang, Liangjie
    Lin, Qingwei
    SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL, 2021, : 2268 - 2272
  • [38] A Novel Use of Reinforcement Learning for Elevated Click-Through Rate in Online Advertising
    Haider, Umair
    Yildiz, Beytullah
    2023 INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND COMPUTATIONAL INTELLIGENCE, CSCI 2023, 2023, : 64 - 70
  • [39] Click-through Rate Prediction and Uncertainty Quantification Based on Bayesian Deep Learning
    Wang, Xiaowei
    Dong, Hongbin
    ENTROPY, 2023, 25 (03)
  • [40] Learning the Click-Through Rate for Rare/New Ads from Similar Ads
    Dave, Kushal
    Varma, Vasudeva
    SIGIR 2010: PROCEEDINGS OF THE 33RD ANNUAL INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH DEVELOPMENT IN INFORMATION RETRIEVAL, 2010, : 897 - 898