Adaptive Designs for Optimizing Online Advertisement Campaigns

  • Andrey PepelyshevEmail author
  • Yuri Staroselskiy
  • Anatoly Zhigljavsky
Conference paper
Part of the Contributions to Statistics book series (CONTRIB.STAT.)


We investigate the problem of adaptive targeting for real-time bidding in online advertisement using independent advertisement exchanges. This is a problem of making decisions based on information extracted from large data sets related to previous experience. We describe an adaptive strategy for optimizing the click through rate which is a key criterion used by advertising platforms to measure the efficiency of an advertisement campaign. We also provide some results of statistical analysis of real data.


User Agent Advertising Campaign Online Advertisement Sequential Splitting Click Through Rate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



The paper is a result of collaboration of Crimtan, a provider of proprietary ad technology platforms and the University of Cardiff. Research of the third author was supported by the Russian Science Foundation, project No. 15-11-30022 “Global optimization, supercomputing computations, and application”.


  1. 1.
    Aly, M., Hatch, A., Josifovski, V., Narayanan, V.K.: Web-scale user modeling for targeting. In: Proceedings of the 21st International Conference Companion on World Wide Web, pp. 3–12. ACM, New York (2012)Google Scholar
  2. 2.
    Box, G.E.P., Wilson, K.B.: On the experimental attainment of optimum conditions (with discussion). J. R. Stat. Soc. Ser. B 13, 1–45 (1951)zbMATHMathSciNetGoogle Scholar
  3. 3.
    eMarketer: US programmatic ad spend tops $10 Billion this year, to double by 2016. (2014)
  4. 4.
    Ermakov, S.M., Zhigljavsky, A.A.: Mathematical Theory of Optimal Design. Nauka, Moscow (1987)Google Scholar
  5. 5.
    Fedorov, V.V., Hackl, P.: Model-Oriented Design of Experiments. Springer, New York (2012)zbMATHGoogle Scholar
  6. 6.
    Fedorov, V.V., Leonov, S.L.: Optimal Design for Nonlinear Response Models. CRC Press, Boca Raton (2013)zbMATHGoogle Scholar
  7. 7.
    Friedman, J.H.: Greedy function approximation: a gradient boosting machine. Ann. Stat. 29, 1189–1232 (2001)CrossRefzbMATHMathSciNetGoogle Scholar
  8. 8.
  9. 9.
  10. 10.
    Jansen, B.J., Mullen, T.: Sponsored search: an overview of the concept, history, and technology. Int. J. Electron. Bus. 6, 114–131 (2008)CrossRefGoogle Scholar
  11. 11.
    McMahan, H.B.: Follow-the-regularized-leader and mirror descent: equivalence theorems and l1 regularization. In: International Conference on Artificial Intelligence and Statistics, Fort Lauderdale, pp. 525–533 (2011)Google Scholar
  12. 12.
    McMahan, H.B., Holt, G., Sculley, D., et al.: Ad click prediction: a view from the trenches. In: Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Chicago, pp. 1222–1230. ACM (2013)Google Scholar
  13. 13.
    Nicholls, S., Malins, A., Horner, M.: Real-time bidding in online advertising. (2014)
  14. 14.
    Rendle, S.: Factorization machines. In: IEEE 10th International Conference on Data Mining (ICDM 2010), Sydney, pp. 995–1000 (2010)Google Scholar
  15. 15.
    Tu, S., Lu, C.: Topic-based user segmentation for online advertising with latent Dirichlet allocation. In: Advanced Data Mining and Applications, pp. 259–269. Springer, Berlin/Heidelberg (2010)Google Scholar
  16. 16.
    Yang, S., Ghose, A.: Analyzing the relationship between organic and sponsored search advertising: positive, negative or zero interdependence? Mark. Sci. 29, 602–623 (2010)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Andrey Pepelyshev
    • 1
    Email author
  • Yuri Staroselskiy
    • 2
  • Anatoly Zhigljavsky
    • 3
  1. 1.Cardiff UniversityCardiffUK
  2. 2.Crimtan, 1 Castle LaneLondonUK
  3. 3.Lobachevskii State University of Nizhnii NovgorodNizhny NovgorodRussia

Personalised recommendations