Skip to main content
SpringerLink
Log in

Data Mining

  • Chapter
  • First Online:
Customer Relationship Management

Part of the book series: Springer Texts in Business and Economics

  • 16k Accesses

Abstract

This chapter describes the importance and benefits of data mining and gives a detailed overview of the underlying process. The data-mining procedure breaks down into five subsections: defining the business objectives, getting the raw data, identifying relevant variables, gaining customer insight, and acting. The discussion of these steps will help the reader understand the overall process of data mining. Furthermore, the process steps are illustrated with the case study of Credite Est (name disguised), a French mid-tier bank. Finally, the case study, “Yapi Kredi—Predictive Model–Based cross-sell Campaign,” shows a comprehensive application of data mining.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Notes

  1. 1.

    An example would be the expectation maximization algorithm (EM) that takes into account the correlation of the data field for which a nonmissing value is to be generated with other data fields.

  2. 2.

    Sometimes, obstacles such as lacking authorization of the data mining team for accessing the required data might emerge. Data miners frequently work with data which other business departments do not have access to. There is a high level of secrecy and trust involved.

  3. 3.

    Kohonen networks belong to the family of neural network techniques. These are powerful data modeling tools able to capture and represent complex input/output relationships for example in target marketing, financial forecasting, or process control. In particular, the objective of a Kohonen network is to generate, out of complex input patterns of arbitrary dimension, a simplified (discrete) map with very few dimensions, say 1 or 2. Thus, the Kohonen network is an approach to quickly understand complex data as a result of a simplification of the structure. For a good overview of neural networks and Kohonen networks please refer to: Principe, Euliano and Lefebvre (2000).

References

  • Olson, J. (2003). Data quality – The accuracy dimension. Amsterdam: Kaufmann.

    Google Scholar 

  • Principe, J. C., Euliano, N. R., & Lefebvre, W. C. (2000). Neural and adaptive systems: Fundamentals through simulations. New York: Wiley.

    Google Scholar 

Download references

Acknowledgments

We thank Frank Block, Ph.D., of FinScore Corporation (Switzerland) for his collaboration on this chapter.

Author information

Authors and Affiliations

  1. J. Mack Robinson College of Business Center for Excellence in Brand and Customer Management, Georgia State University, Atlanta, Georgia, USA

    Prof. V. Kumar

  2. Department of Retailing and Customer Management, University of Cologne, Köln, Germany

    Werner Reinartz

Authors
  1. Prof. V. Kumar
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Werner Reinartz
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to V. Kumar .

Rights and permissions

Reprints and permissions

Copyright information

© 2012 Springer-Verlag Berlin Heidelberg

About this chapter

Cite this chapter

Kumar, V., Reinartz, W. (2012). Data Mining. In: Customer Relationship Management. Springer Texts in Business and Economics. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-20110-3_7

Download citation

Publish with us

Policies and ethics

Search

Navigation