PREDICT CUSTOMER CHURN THROUGH CLASS IMBALANCE WITH MODIFIED RIPPER ALGORITHM

M. Rajeswari, S. Deepika

Abstract


Competitive advantage is gained by firm through new areas such as data warehousing, data mining, and campaign management software have made Customer Relationship Management (CRM). This research aims to develop methodologies for predicting customer churn in advance, while keeping misclassification rates to a minimum.


Keywords


Churn, Class Imbalance, Customer Relationship Management, Data Mining.

References


Hyafil L, and Rivest RL. (1976). "Constructing optimal binary decision trees is NP complete", Information Processing Letters, Vol.5(1), pp.15-17.

Quinlan JR. (1987). "Simplifying decision trees", International Journal of Man Machine Studies, Vol.27, pp.221-234.

Weiss GM, and Provost F. (2001). "The effect of class distribution on classifier learning: An empirical study", Dept. Comput. Sci., Rutgers Univ., Newark, NJ, Tech. Rep. TR-43.

Weiss GM, and Provost F. (2003). "Learning when training data are costly: The effect of class distribution on tree induction", J. Artif. Intell. Res., Vol.19, pp.315–354.

Zantema H, and Bodlaender HL. (2000). "Finding Small Equivalent Decision Trees is Hard", International Journal of Foundations of Computer Science, Vol.11(2), pp.343-354.


Refbacks

  • There are currently no refbacks.


Send mail to ijmss@ijmss.com with questions or comments about this web site. 

International Journal of Management and Social Sciences, All rights reserved.