Header logo is ei

Data mining problems and solutions for response modeling in CRM

2006

Article

ei


We present three data mining problems that are often encountered in building a response model. They are robust modeling, variable selection and data selection. Respective algorithmic solutions are given. They are bagging based ensemble, genetic algorithm based wrapper approach and nearest neighbor-based data selection in that order. A real world data set from Direct Marketing Educational Foundation, or DMEF4, is used to show their effectiveness. Proposed methods were found to solve the problems in a practical way.

Author(s): Cho, S. and Shin, H. and Yu, E. and Ha, K. and MacLachlan, D.
Journal: Entrue Journal of Information Technology
Volume: 5
Number (issue): 1
Pages: 55-64
Year: 2006
Month: March
Day: 0

Department(s): Empirical Inference
Bibtex Type: Article (article)

Digital: 0
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF

BibTex

@article{3816,
  title = {Data mining problems and solutions for response modeling in CRM},
  author = {Cho, S. and Shin, H. and Yu, E. and Ha, K. and MacLachlan, D.},
  journal = {Entrue Journal of Information Technology},
  volume = {5},
  number = {1},
  pages = {55-64},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  month = mar,
  year = {2006},
  month_numeric = {3}
}