Research on advertising click-through rate estimation based on feature learning

被引：0

作者：

Zhang Z.-Q. ^{[1
]}

Zhou Y. ^{[1
]}

Xie X.-Q. ^{[1
]}

Pan H.-W. ^{[1
]}

机构：

[1] College of Computer Science and Technology, Harbin Engineering University, Harbin

来源：

Jisuanji Xuebao/Chinese Journal of Computers | 2016年 / 39卷 / 04期

基金：

中国国家自然科学基金;

关键词：

Click through rate; Computational advertising; Deep learning; Social media; Social networks; Sponsored search; Tensor decomposition;

D O I：

10.11897/SP.J.1016.2016.00780

中图分类号：

学科分类号：

摘要：

The issue of click through rate estimation in sponsored search has been widely studied in information retrieval, machine learning and query recommendation etc. Some related studies, such as the methods in which features are obtained by setting the feature extraction scheme or aiming at user behavior modeling, did not take into account those essential characteristics including the sparseness of advertising data and highly nonlinear association between features. In order to fully mining the hidden rules in advertising data, this paper proposes a method that can learn the sparse feature of advertising data. Our method combines dimension reduction based on tensor decomposition and takes full advantage of feature learning to portraying the nonlinear associated relationship of data to solve sparse feature learning problems. Finally, the comparison experiment shows this method has the desired effect of improving the accuracy of CTR estimation. © 2016, Science Press. All right reserved.

引用

页码：780 / 794

页数：14

共 30 条

[1]

Broder A.Z., Computational advertising, Proceedings of the 19th Annual ACM-SIAM Symposium on Discrete Algorithms, SODA, (2008)

[2]

Zhou A.-Y., Zhou M.-Q., Gong X.-Q., Computational advertising: A data-centric comprehensive Web application, Chinese Journal of Computers, 34, 10, pp. 1805-1819, (2011)

[3]

Richardson M., Dominowska E., Ragno R., Predicting clicks: Estimating the click-through rate for new ads, Proceedings of the 16th International Conference on World Wide Web, pp. 521-530, (2007)

[4]

Srikant R., Basu S., Wang N., Et al., User browsing models: Relevance versus examination, Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 223-232, (2010)

[5]

Hillard D., Schroedl S., Manavoglu E., Et al., Improving ad relevance in sponsored search, Proceedings of the 3rd ACM International Conference on Web Search and Data Mining, pp. 361-370, (2010)

[6]

Dupret G., Liao C., A model to estimate intrinsic document relevance from the clickthrough logs of a web search engine, Proceedings of the 3rd ACM International Conference on Web Search and Data Mining, pp. 181-190, (2010)

[7]

Dupret G.E., Piwowarski B., A user browsing model to predict search engine click data from past observations, Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 331-338, (2008)

[8]

Guo F., Liu C., Kannan A., Et al., Click chain model in web search, Proceedings of the 18th International Conference on World Wide Web, pp. 11-20, (2009)

[9]

Chapelle O., Zhang Y., A dynamic Bayesian network click model for web search ranking, Proceedings of the 18th International Conference on World Wide Web, pp. 1-10, (2009)

[10]

Zhang W.V., Jones R., Comparing click logs and editorial labels for training query rewriting, Proceedings of the WWW 2007 Workshop on Query Log Analysis: Social and Technological Challenges, (2007)

← 1 2 3 →