Targeted Display Advertising using Machine Learning
DOI:
https://doi.org/10.59890/ijaamr.v1i3.715Keywords:
Problem formulation, Multi-stage Architecture, Large-scale Machine learning system, Experimental ValidationAbstract
This paper delves into the intricate challenges of problem formulation and data representation in the context of a large-scale machine learning system for targeted display advertising. Unlike traditional models, this system is not just conceptual but has been operational for years across thousands of advertising campaigns. Since obtaining ideal training data is cost-prohibitive, data is sourced from related domains and tasks and then adapted for the target task. The paper outlines the architecture of this multi-stage transfer learning system, emphasizing the problem formulation aspects. Extensive experiments demonstrate the value of each transfer stage. Real-world results with diverse advertising clients from various industries showcase the system's performance. The paper concludes with valuable insights gained from over half a decade of work on this complex, widely deployed machine learning system.References
Agarwal,D.,Agrawal,R.,Khanna,R.,Kota,N.:Estimatingratesofrareeventswithmultiplehierarchiesthroughscal- able log-linear models. In: Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 213–222 (2010)
Attenberg, J., Ipeirotis, P., Provost, F.: Beat the machine: Challenging workers to find the unknown unknowns. In: Workshops at the Twenty-Fifth AAAI Conference on Artificial Intelligence (2011)
Attenberg,J.,Provost,F.:Whylabelwhenyoucansearch?strategiesforapplyinghumanresourcestobuildclassifica- tion models under extreme class imbalance. In: KDD (2010)
Bottou, L.: Large-scale machine learning with stochastic gradient descent. In: Y. Lechevallier, G. Saporta (eds.) Pro- ceedings of the 19th International Conference on Computational Statistics (COMPSTAT’2010), pp. 177–187. Springer, Paris, France (2010). URL http://leon.bottou.org/papers/bottou-2010
Breiman,L.:Stackedregressions.Machinelearning24(1),49–64(1996)
Chen, Y., Pavlov, D., Canny, J.: Large-scale behavioral targeting. In: Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 209–218. ACM (2009)
Dalessandro,B.,Hook,R.,Perlich,C.,Provost,F.:Evaluatingandoptimizingonlineadvertising:Forgettheclick,but there are good proxies. NYU Working Paper CBA-12-02 (2012)
Evgeniou, T., Pontil, M.: Regularized multi–task learning. In: Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 109–117. ACM (2004)
Fawcett,T.,Provost,F.:Adaptivefrauddetection.Dataminingandknowledgediscovery1(3),291–316(1997)
Stitelman, O., Dalessandro, B., Perlich, C., Provost, F.: Estimating the effect of online display advertising on browser conversion. Data Mining and Audience Intelligence for Advertising (ADKDD 2011) p. 8 (2011)
TheNewYorkTimes:Your online attention, boughtinaninstant(2012)
Weinberger,K.,Dasgupta,A.,Langford,J.,Smola,A.,Attenberg,J.:Featurehashingforlargescalemultitasklearning. In: Proceedings of the 26th Annual International Conference on Machine Learning, pp. 1113–1120. ACM (2009)
Weiss,G.,Provost,F.:Learningwhentrainingdataarecostly:Theeffectofclassdistributionontreeinduction.J.Artif. Intell. Res. (JAIR) 19, 315–354 (2003)
Xue, Y., Liao, X., Carin, L., Krishnapuram, B.: Multi-task learning for classification with dirichlet process priors. The Journal of Machine Learning Research 8, 35–63 (2007)
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Chaitali Khachane
This work is licensed under a Creative Commons Attribution 4.0 International License.
