Churn prediction has become a significant requirement for all customer centric organizations. Accurate prediction of churn can effectively improve customer loyalty and improve profits for the organization. This work presents an effective model that uses a combination of ensemble learning and minimax probability machines to provide a churn prediction system. The model has its major focus towards improving the profitability of the organization. The ensemble learning model has been designed to be computationally efficient, while the weight factors used in the minimax probability machines ensures reduction in losses, hence ensuring profitability. Experiments were performed and comparisons with existing models indicates that the model shows high performance, with 8% improved accuracy levels, indicating improved churn predictions.
V Jude Nirmal
St. Joseph’s College, India
Churn prediction, Ensemble learning, Minimax Probability Machines, Extra Trees Classifier, Profit Maximization
| January | February | March | April | May | June | July | August | September | October | November | December |
| 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 12 , Issue: 1 , Pages: 2469-2473 )
Date of Publication :
October 2021
Page Views :
327
Full Text Views :
1
|