Abstract
The advertisement logs accumulated in the Internet have problems such as sparse data, large feature quantity, and extremely uneven distribution of positive and negative samples, which made it difficult to obtain interesting features and to improve precision for single prediction models. In response to these problems, this paper proposes a CTR prediction model based on GBDT-Stacking. GBDT-Stacking model uses the GBDT to automatically extract and transform features suitable for prediction and uses Stacking model to predict CTR of user, which improves the performance of baseline effectively. The experimental results in the real advertising dataset show that the GBDT-Stacking model of this paper uses increased by at least 4% compared to single model in AUC value.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Gao, C., Lu, Z.M.: Developing tendency and characteristics of online advertisement[J]. J. Harbin Inst. Technol. (Soc. Sci. Ed.) 02, 122–125 (2003)
Zhou, A.Y., Zhou, M.Q., Gong, X.Q.: Computational advertising: a data-centric comprehensive web application[J]. Chin. J. Comput. 34(10), 1805–1819 (2011)
Richardson, M., Dominowska, E., Ragno, R.: Predicting clicks: estimating the click-through rate for new ads[C]. In: International Conference on World Wide Web, pp. 8–12 (2007)
Shen, F.Y., Dai, G.J., Dai, C.L., et al.: CTR prediction for online advertising based on a features conjunction model. J. Tsinghua Univ. (Sci. Technol.) 58(04), 374–379 (2018)
Rendle, S.: Factorization machines with libFM. ACM Trans. Intell. Syst. Technol. 3(3), 1–22 (2012)
Juan, Y., Zhuang, Y., Chin, W.S., et al.: Field-aware factorization machines for CTR Prediction. In: RecSys 2016 Proceedings of the 10th ACM Conference on Recommender Systems, 15. 9. 2016–19. 9. 2016, Boston, Massachusetts, USA, pp. 43–50. ACM Press (2016)
Zhang, Y., Dai, H., Xu, C., et al.: Sequential click prediction for sponsored search with recurrent neural networks (2014)
Zhang, Z.Q., Zhou, Y., Xie, X.Q., et al.: Research on advertising click-through rate estimation based on feature learning. Chin. J. Comput. 39(04), 780–794 (2016)
Dave, K.S., Varma, V.: Learning the click-through rate for rare/new ads from similar ads. In: International ACM SIGIR Conference on Research & Development in Information Retrieval (2010)
Yue, Q., Wang, C.L., Zhu, Y.L., et al.: Click-through rate prediction of online advertisements based on probabilistic graphical model. J. East China Normal Univ. (Nat. Sci.) 03, 15–25 (2013)
Qin, C.Y.: Research on influencing factors of keyword conversion rate based on hierarchical Bayesian method. Stat. Decis. 19, 64–67 (2016)
Xiao, Y., Bi, J.F., Han, Y., et al.: Click rate prediction research in online advertising. J. East China Normal Univ. (Nat. Sci.) 05, 80–86 (2017)
Yang, C.C., Mei, J.J., Wu, Y., et al.: Advertising click-through rate prediction based on feature dimension reduction and deep belief network. Comput. Eng. Des. 39(12), 3700–3704 (2018)
She, X.Y., Wang, S.P.: Research on advertising click through rate prediction model based on CNN-LSTM network. Comput. Eng. Appl. 55(02), 193–197 (2019)
He, X., Pan, J., Jin, O., et al.: Practical lessons from predicting clicks on ads at facebook. In: Proceedings of 20th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pp. 1–9. ACM (2014)
Friedman, J.H.: Greedy function approximation: a gradient boosting machine. Ann. Stat. 29(5), 1189–1232 (2001)
Wang, Z.N., Xiao, W.M., Wang, J.: A practical pipeline with stacking models for KKBOX’s churn prediction challenge. In: Proceedings of the 11th ACM International Conference on Web Search and Data Mining (2018)
Tian, C.L., Zhang, X., Pan, B., Yang, C., Xu, Y.R.: Research and implementation of feature extraction methods on Internet CTR prediction model. Appl. Res. Comput. 34(2), 334–338 (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
He, X., Pan, W., Cheng, H. (2020). Research on Advertising Click-Through Rate Prediction Model Based on Ensemble Learning. In: Han, H., Wei, T., Liu, W., Han, F. (eds) Recent Advances in Data Science. IDMB 2019. Communications in Computer and Information Science, vol 1099. Springer, Singapore. https://doi.org/10.1007/978-981-15-8760-3_6
Download citation
DOI: https://doi.org/10.1007/978-981-15-8760-3_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-8759-7
Online ISBN: 978-981-15-8760-3
eBook Packages: Computer ScienceComputer Science (R0)