Soft Frequency Capping for Improved Ad Click Prediction in Yahoo Gemini Native

被引:10
作者
Aharon, Michal [1 ]
Kaplan, Yohay [1 ]
Levy, Rina [1 ]
Somekh, Oren [1 ]
Blanc, Ayelet [2 ]
Eshel, Neetai [2 ]
Shahar, Avi [2 ]
Singer, Assaf [2 ]
Zlotnik, Alex [2 ]
机构
[1] Yahoo Res, Haifa, Israel
[2] Tech Yahoo, Tel Aviv, Israel
来源
PROCEEDINGS OF THE 28TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT (CIKM '19) | 2019年
关键词
Recommendation systems; collaborative filtering; ad click-prediction; ad-ranking; ad fatigue; soft frequency capping;
D O I
10.1145/3357384.3357801
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Yahoo's native advertising (also known as Gemini native) serves billions of ad impressions daily, reaching a yearly run-rate of many hundred of millions USD. Driving the Gemini native models that are used to predict both click probability (pCTR) and conversion probability (pCONV) is Offset - a feature enhanced collaborative-filtering (CF) based event prediction algorithm. Offset is a one-pass algorithm that updates its model for every new batch of logged data using a stochastic gradient descent (SGD) based approach. Since Offset represents its users by their features (i.e., user-less model) due to sparsity issues, rule based hard frequency capping (HFC) is used to control the number of times a certain user views a certain ad. Moreover, related statistics reveal that user ad fatigue results in a dramatic drop in click through rate (CTR). Therefore, to improve click prediction accuracy, we propose a soft frequency capping (SFC) approach, where the frequency feature is incorporated into the Offset model as a user-ad feature and its weight vector is learned via logistic regression as part of Offset training. Online evaluation of the soft frequency capping algorithm via bucket testing showed a significant 7.3% revenue lift. Since then, the frequency feature enhanced model has been pushed to production serving all traffic, and is generating a hefty revenue lift for Yahoo Gemini native. We also report related statistics that reveal, among other things, that while users' gender does not affect ad fatigue, the latter seems to increase with users' age.
引用
收藏
页码:2793 / 2801
页数:9
相关论文
共 19 条
[1]  
Aharon M., 2013, P RECSYS
[2]   Adaptive Online Hyper-Parameters Tuning for Ad Event-Prediction Models [J].
Aharon, Michal ;
Kagian, Amit ;
Somekh, Oren .
WWW'17 COMPANION: PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, 2017, :672-679
[3]  
Aizenberg N., 2012, P 21 INT C WORLD WID, P1
[4]  
[Anonymous], 2013, RECSYS 2013
[5]  
Bell RM., 2007, Acm Sigkdd Explorations Newsletter, V9, P75, DOI [10.1145/1345448.1345465, DOI 10.1145/1345448.1345465]
[6]   Frequency capping in online advertising [J].
Buchbinder, Niv ;
Feldman, Moran ;
Ghosh, Arpita ;
Naor, Joseph .
JOURNAL OF SCHEDULING, 2014, 17 (04) :385-398
[7]  
Dean J, 2004, USENIX ASSOCIATION PROCEEDINGS OF THE SIXTH SYMPOSIUM ON OPERATING SYSTEMS DESIGN AND IMPLEMENTATION (OSDE '04), P137
[8]  
Duchi J, 2011, J MACH LEARN RES, V12, P2121
[9]   Internet advertising and the generalized second-price auction: Selling billions of dollars worth of keywords [J].
Edelman, Benjamin ;
Ostrovsky, Michael ;
Schwarz, Michael .
AMERICAN ECONOMIC REVIEW, 2007, 97 (01) :242-259
[10]   An introduction to ROC analysis [J].
Fawcett, Tom .
PATTERN RECOGNITION LETTERS, 2006, 27 (08) :861-874