基于序列特征的点击率预测模型

doi:10.3969/j.issn.1000-5641.201921006

摘要/Abstract

摘要： 点击率预测模型是主流推荐系统中十分重要的部分. 根据点击率预测的打分来调整商品的展示策略, 对提高业务的转化率、改进用户体验等有着重要的意义. 传统的点击率预测模型是利用用户特征和商品特征, 对点击率进行预测. 然而, 用户行为序列的结构特征, 如周期性规律、趋势等也能一定程度地体现用户行为的倾向. 针对部分信息利用上的空缺, 使用时间序列分析单元, 将提取用户行为序列的特征作为用户特征的扩展, 结合因子分解机结构将其与用户、商品特征进行交叉, 能够有效提高特征质量, 优化点击率预测模型的性能. 实验表明, 结合用户行为序列特征进行交叉优化的方法能够对点击率预测模型的表现带来很大提升, 提高点击率预测的精度.

关键词: 点击率预测, 推荐系统, 自回归滑动平均模型, 因子分解机

Abstract: The click-through rate (CTR) prediction model is an important component of mainstream recommendation systems. The model assigns a score to recommended items according to the predicted CTR and generates an optimized scoring function which in-turn influences an item’s display strategy; this helps generate improved business conversion rates and a better user experience. Generally, CTR prediction models utilize both user and item features to predict CTR. However, structural characteristics of user behavior, such as frequency and trends, can also reflect behavioral tendencies. Given the absence of this information, this paper analyzes user behavior sequences as a time series and extracts latent features. Factorization machines are then used to learn from user/item features combined with sequence features to improve the quality of prediction. Experiments show that the sequence feature-based methods improve the performance of CTR prediction models and make CTR prediction more accurate.

Key words: click-through rate (CTR) prediction, recommendation system, auto-regressive moving average model（ARMA）, factorization machine

中图分类号:

TP391

朱思涵, 浦剑. 基于序列特征的点击率预测模型[J]. 华东师范大学学报（自然科学版）, 2020, 2020(4): 134-146.

ZHU Sihan, PU Jian. Model for click-through rate prediction based on sequence features[J]. Journal of East China Normal University(Natural Science), 2020, 2020(4): 134-146.

参考文献

[1] 纪文迪, 王晓玲, 周傲英. 广告点击率估算技术综述 [J]. 华东师范大学学报(自然科学版), 2013(3): 2-14
[2] ZHAO Z D, SHANG M S. User-based collaborative-filtering recommendation algorithms on hadoop [C]// 2010 3rd International Conference on Knowledge Discovery and Data Mining. IEEE, 2010: 478-481.
[3] PIRASTEH P, JUNG J J, HWANG D. Item-based collaborative filtering with attribute correlation: A case study on movie recommendation [C]// Asian Conference on Intelligent Information and Database Systems 2014: Intelligent Information and Database Systems. Cham: Springer, 2014: 245-252. DOI: 10.1007/978-3-319-05458-2_26.
[4] RICHARDSON M, DOMINOWSKA E, RAGNO R J, et al. Predicting clicks: Estimating the click-through rate for new ads [C]// Proceedings of the 16th International Conference on World Wide Web. ACM, 2007: 521-530. DOI: 10.1145/1242572.1242643.
[5] JOACHIMS T. Optimizing search engines using clickthrough data [C]//Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2002: 133-142.
[6] ZHANG W N, DU T M, WANG J. Deep learning over multi-field categorical data [C]//European Conference on Information Retrieval 2016: Advances in Information Retrieval. Cham: Springer, 2016: 45-57. DOI: 10.1007/978-3-319-30671-1_4.
[7] RENDLE S. Factorization machines [C]//2010 IEEE International Conference on Data Mining. IEEE, 2010: 995-1000. DOI: 10.1109/ICDM.2010.127.
[8] QUINLAN J R. Induction of decision trees [J]. Machine Learning, 1986, 1(1): 81-106. DOI: 10.1023/A:1022643204877.
[9] SCHAPIRE R E. A brief introduction to boosting [C]// Proceedings of the 16th International Joint Conference on Artificial Intelligence. San Francisco: Morgan Kaufmann Publishers Inc, 1999: 1401-1406.
[10] CHEN T, GUESTRIN C. Xgboost: A scalable tree boosting system [C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2016: 785-794.
[11] QU Y, CAI H, REN K, et al. Product-based neural networks for user response prediction [C]//2016 IEEE 16th International Conference on Data Mining (ICDM). IEEE, 2016: 1149-1154.
[12] CHENG H T, KOC L, HARMSEN J, et al. Wide & deep learning for recommender systems [C]//Proceedings of the 1st Workshop on Deep Learning for Recommender Systems. ACM, 2016: 7-10.
[13] GUO H, TANG R, YE Y, et al. DeepFM: A factorization-machine based neural network for CTR prediction [C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence. AAAI, 2017: 1725-1731.
[14] AGUIAR E, NAGRECHA S, CHAWLA N V. Predicting online video engagement using clickstreams [C]//2015 IEEE International Conference on Data Science and Advanced Analytics (DSAA). IEEE, 2015: 1-10.
[15] 李思琴, 林磊, 孙承杰. 基于卷积神经网络的搜索广告点击率预测 [J]. 智能计算机与应用, 2015(5): 22-25. DOI: 10.3969/j.issn.2095-2163.2015.05.007
[16] HE X R, PAN J F, JIN O, et al. Practical lessons from predicting clicks on ads at Facebook [C]//ADKDD’14: Proceedings of the 8th International Workshop on Data Mining for Online Advertising. ACM, 2014: pp.1-9. DOI: 10.1145/2648584.2648589.
[17] 叶健, 赵慧. 基于大规模弹幕数据监听和情感分类的舆情分析模型 [J]. 华东师范大学学报(自然科学版), 2019(3): 86-100
[18] BROCKWELL P J, DAVIS R A, CALDER M V. Introduction to Time Series and Forecasting [M]. New York: Springer, 2002: 73-96.
[19] XIAO J, YE H, HE X N, et al. Attentional factorization machines: Learning the weight of feature interactions via attention networks [C]//Proceedings of the 26th International Joint Conference on Artificial Intelligence. AAAI, 2017: 3119-3125.