A Non-sequential Representation of Sequential Data for Churn Prediction
We investigate the length of event sequence giving best predictions when using a continuous HMM approach to churn prediction from sequential data. Motivated by observations that predictions based on only the few most recent events seem to be the most accurate, a non-sequential dataset is constructed from customer event histories by averaging features of the last few events. A simple K-nearest neighbor algorithm on this dataset is found to give significantly improved performance. It is quite intuitive to think that most people will react only to events in the fairly recent past. Events related to telecommunications occurring months or years ago are unlikely to have a large impact on a customer’s future behaviour, and these results bear this out. Methods that deal with sequential data also tend to be much more complex than those dealing with simple non-temporal data, giving an added benefit to expressing the recent information in a non-sequential manner.
KeywordsHide Markov Model Event History Recent Event Combination Method Hide State
Unable to display preview. Download preview PDF.
- 1.Bilmes, J.: A gentle tutorial on the em algorithm and its application to parameter estimation for gaussian mixture and hidden markov models (1997)Google Scholar
- 6.Freund, Y., Schapire, R.E.: Experiments with a new boosting algorithm. In: Proceedings of the 13th International Conference on Machine Learning, pp. 148–156. Morgan Kaufmann, San Francisco (1996)Google Scholar
- 7.Haddon, J., Tiwari, A., Roy, R., Ruta, D.: Churn prediction: Does technology matter (2006)Google Scholar
- 9.Murphy, K.: A hmm toolbox for matlab, http://www.cs.ubc.ca/~murphyk/software/hmm/hmm.html
- 10.Quinlan, J.R.: C4.5: programs for machine learning. Morgan Kaufmann Publishers Inc., San Francisco (1993)Google Scholar
- 13.Yan, L., Miller, D.J., Mozer, M.C., Wolniewicz, R.: Improving prediction of customer behaviour in non-stationary environments. In: Proc. of Int. Joint Conf. on Neural Networks, pp. 2258–2263 (2001)Google Scholar