Soft Computing

, Volume 21, Issue 3, pp 651–665

A methodology based on Deep Learning for advert value calculation in CPM, CPC and CPA networks

  • Luis Miralles-Pechuán
  • Dafne Rosso
  • Fernando Jiménez
  • Jose M. García

DOI: 10.1007/s00500-016-2468-4

Cite this article as:
Miralles-Pechuán, L., Rosso, D., Jiménez, F. et al. Soft Comput (2017) 21: 651. doi:10.1007/s00500-016-2468-4


In this research, we propose a methodology for advert value calculation in CPM, CPC and CPA networks. Accurately estimating this value increases the three previous networks’ incomes by selecting the most profitable advert. By increasing income, publishers are better paid and improved services are afforded to advertisers. To develop this methodology, we propose a system based on traditional Machine Learning methods and Deep Learning methods. The system has two inputs and one output. The inputs are the user visit and the data about the advertiser. The output is the advert value expressed in dollars. Deep Learning predicts model behavior more precisely for many supervised problems. The three experiments carried out allow us to conclude that DL is a supervised method that is very efficient in the classification of spam adverts and in the estimation of the CTR. In the prediction of online sales, DLNN have shown, on average, worse performance than cubist and random forest methods, although better performance than model tree, model rules and linear regression methods.


Advertisement value calculation in CPM, CPC and CPA networks Deep Learning methods in online advertising Sales prediction Spam probability calculation CTR estimation Deep Learning in advertisement value calculation 

Funding information

Funder NameGrant NumberFunding Note
  • TIN2013-45491-R
  • TIN2015-66972-C5-3-R

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  1. 1.Universidad Panamericana, Campus México, Facultad de IngenieríaCiudad de MéxicoMéxico
  2. 2.Faculty of Computer ScienceUniversidad de MurciaMurciaSpain

Personalised recommendations