Segmentation-based modeling for advanced targeted marketing

ABSTRACT

Fingerhut Business Intelligence (BI) has a long and successful history of building statistical models to predict consumer behavior. The models constructed are typically segmentation-based models in which the target audience is split into subpopulations (i.e., customer segments) and individually tailored statistical models are then developed for each segment. Such models are commonly employed in the direct-mail industry; however, segmentation is often performed on an ad-hoc basis without directly considering how segmentation affects the accuracy of the resulting segment models. Fingerhut BI approached IBM Research with the problem of how to build segmentation-based models more effectively so as to maximize predictive accuracy. The IBM Advanced Targeted Marketing-Single Events^TM (IBM ATM-SE^TM) solution is the result of IBM Research and Fingerhut BI directing their efforts jointly towards solving this problem. This paper presents an evaluation of ATM-SE's modeling capabilities using data from Fingerhut's catalog mailings.

REFERENCES

• 1: Anthes, G., Optimal Results. Computerworld, 2000. 34(47): p. 60-61.
• 2: Chidanand Apte , Edna Grossman , Edwin P. D. Pednault , Barry K. Rosen , Fateh A. Tipu , Brian White, Probabilistic Estimation-Based Data Mining for Discovering Insurance Risks, IEEE Intelligent Systems, v.14 n.6, p.49-58, November 1999 [doi>10.1109/5254.809568]
• 3: Bibelnieks, E. and D. Campbell, Mail Stream Streamlining. Catalog Age, 2000. 17(12): p. 118-120.
• 4: Breiman, L., et al., Classification and Regression Trees. 1984, Monterrey, CA.: Wadsworth.
• 5: Deb Campbell , Randy Erdahl , Doug Johnson , Eric Bibelnieks , Michael Haydock , Mark Bullock , Harlan Crowder, Optimizing Customer Mail Streams at Fingerhut, Interfaces, v.31 n.1, p.77-90, January 2001 [doi>10.1287/inte.31.1.77.9691]
• 6: Jonathan R. M. Hosking , Edwin P. D. Pednault , Madhu Sudan, A statistical perspective on data mining, Future Generation Computer Systems, v.13 n.2-3, p.117-134, Nov. 1997 [doi>10.1016/S0167-739X(97)00016-2]
• 7: Lachs, J., Data Mining Digs In. American Demographics, 1999.21(7): p. 3845.
• 8: C. J. Matheus , P. K. Chan , G. Piatetsky-Shapiro, Systems for Knowledge Discovery in Databases, IEEE Transactions on Knowledge and Data Engineering, v.5 n.6, p.903-913, December 1993 [doi>10.1109/69.250073]
• 9: Natarajan, R. and E.P.D. Pednault, Decision Trees with Node Regression Estimators for Massive Data Sets. 2001. IBM Research Report (in preparation)
• 10: Natarajan, R. and E.P.D. Pednault. Using Simulated Pseudo Data To Speed Up Statistical Predictive Modeling From Massive Data Sets. in First SIAM International Conference on Data Mining. 2001. Chicago.
• 11: Pednault, E.P.D., Statistical Learning Theory, in The MIT Encyclopedia of the Cognitive Sciences, R.A. Wilson and F.C. Keil, Editors. 1998, MIT Press: Cambridge, MA. p. 798-801.
• 12: J. Ross Quinlan, C4.5: programs for machine learning, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1993
• 13: Staff_Reporter, Mail Order Giant Uses Business Intelligence to Make Every Mailing Count. KM World, 1999.8(5).
• 14: Wreden, N., Making Marketing Personal. Beyond Computing, 1999: p. 24-25.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications

General Terms:
Algorithms, Measurement, Performance

Keywords:
Segmentation-based models, decision trees, feature selection, linear regression, logistic regression, targeted marketing