Advertisement

MLP modeling for search advertising price prediction

  • Hyunhee ParkEmail author
Original Research
  • 13 Downloads

Abstract

As the use of online and various smart devices spread, the use of online search engines became more active. As Internet shopping has evolved through online search engines, competition is under way to launch its link at the top of search engines to expose its links to prospective shoppers. This trend has contributed to the increase in advertising costs in the search advertising market. In this case, the value of the search keyword is generally calculated based on the frequency of the search keyword, however the search engine configures the price of the search keyword through the private auction method without disclosing the price in real time. Finally, it is difficult to reach the exact price and position by passive statistical method in order to predict the price of the search keyword. There is a growing demand for automation methodologies to perform this process quickly and efficiently. In this paper, we propose a Multi-Layer Perceptron (MLP) Neural Network modeling method that estimates bid prices of search keywords by collecting search keywords. MLP is used because it uses generalized delta learning rules and easily gets trained in less number of iterations. In this paper, we propose a MLP based prediction modeling to predict optimal bidding price of the keyword in a specific ranking of search engine.

Keywords

Prediction modeling Deep learning MLP Keyword ad 

Notes

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2017R1C1B5017556).

References

  1. Auerbach J, Galenson J, Sundararajan M (2008) In: Proceedings an empirical analysis of return on investment maximization in sponsored search auctions, International workshop data mining audience Intell Ad ADKDD08Google Scholar
  2. Brooks N (2004) In: Proceedings The Atlas rank report: how search engine rank impacts traffic. Insights, Atlas Institute Digital MarketingGoogle Scholar
  3. Gopal R, Li X, Sankaranarayanan R (2011) Online keyword based advertising impact of ad impressions on own channel and cross channel click through rates. Decis Support Syst 52:1–31CrossRefGoogle Scholar
  4. Graepel T, Candela J, Borchert T, Herbrich R (2010) In: Proceedings Web-scale Bayesian click through rate prediction for sponsored search advertising in Microsoft Bing search engine, International Conference on Machine Learning. ICMLGoogle Scholar
  5. Hou L (2015) A hierarchical bayesian network-based approach to keyword auction. IEEE Trans Eng Manag 62:217–225CrossRefGoogle Scholar
  6. Jerath K, Ma L, Park Y, Srinivasan K (2011) A position Paradox in sponsored search auctions. Mark Sci 30:612–627CrossRefGoogle Scholar
  7. Kingma D, Ba J (2015) In Proceeding ADAM: a method for stochastic optimization, International conference on learning representations. ICLRGoogle Scholar
  8. Lauritzen S (1995) The EM algorithm for graphical association models with missing data. Comput Stat Data Anal 19:191–201CrossRefzbMATHGoogle Scholar
  9. Oliver JR, Randolph EB (2011) From generic to branded: a model of spillover in paid search advertising. J Mark Res 48:87–102CrossRefGoogle Scholar
  10. Shuai Y, Wang J, Zhao X (2013) In: Proceedings Real-time bidding for online advertising: measurement and analysis, International Workshop on Data Mining for Online Advertising. ADKDD13Google Scholar
  11. Stepanchuk T (2008) In: Proceedings an empirical examination of the relation between bids and positions of ads in sponsored search, BLEDGoogle Scholar
  12. Sur C (2018) DeepSeq: learning browsing log data based personalized security vulnerabilities and counter intelligent measures. J. Ambient Intell Hum Comput 1–30Google Scholar
  13. Suto J, Oniga S (2017) Efficiency investigation of artificial neural networks in human activity recognition. J Ambient Intell Hum Comput 9:1049–1060CrossRefGoogle Scholar
  14. Toshitaka M, Kazuki T, Toshihiko W, Akihisa K, Noboru S (2018) Resource propagation algorithm considering predicates to complement knowledge bases in linked data. Int J Space-Based Situat Comput 8:115–121CrossRefGoogle Scholar
  15. Xiaohui L, Yang Z, Hongbin D, Jun H (2016) A novel near-parallel version of k-means algorithm for n-dimensional data objects using MPI. Int J Grid Util Comput 7:80–91CrossRefGoogle Scholar
  16. Yala N, Fergani B, Fleury A (2017) Towards improving feature extraction and classification for activity recognition on streaming data. J Ambient Intell Hum Comput 8:177–189CrossRefGoogle Scholar
  17. Yu-Cheng W (2018) Prediction of engine failure time using principal component analysis, categorical regression tree, and back propagation network. J. Ambient Intell Hum Comput. 1–9Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Computer SoftwareKorean Bible UniversitySeoulSouth Korea

Personalised recommendations