Data Split Strategiesfor Evolving Predictive Models

  • Vikas C. Raykar
  • Amrita Saha
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9284)


A conventional textbook prescription for building good predictive models is to split the data into three parts: training set (for model fitting), validation set (for model selection), and test set (for final model assessment). Predictive models can potentially evolve over time as developers improve their performance either by acquiring new data or improving the existing model. The main contribution of this paper is to discuss problems encountered and propose workflows to manage the allocation of newly acquired data into different sets in such dynamic model building and updating scenarios. Specifically we propose three different workflows (parallel dump, serial waterfall, and hybrid) for allocating new data into the existing training, validation, and test splits. Particular emphasis is laid on avoiding the bias due to the repeated use of the existing validation or the test set.


Data splits Model assessment Predictive models 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
  2. 2.
    Chatfield, C.: Model uncertainty, data mining and statistical inference. Journal of the Royal Statistical Society. Series A (Statistics in Society) 158(3), 419–466 (1995)CrossRefGoogle Scholar
  3. 3.
    Efron, B., Tibshirani, R.: An introduction to the bootstrap. Chapman and Hall (1993)Google Scholar
  4. 4.
    Faraway, J.: Data splitting strategies for reducing the effect of model selection on inference. Computing Science and Statistics 30, 332–341 (1998)Google Scholar
  5. 5.
    Gama, J., Zliobaite, I., Bifet, A., Pechenizkiy, M., Bouchachia, A.: A survey on concept drift adaptation. ACM Computing Surveys 46(4) (2014)Google Scholar
  6. 6.
    Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer Series in Statistics (2009)Google Scholar
  7. 7.
    Kohavi, R.: A study of cross-validation and bootstrap for accuracy estimation and model selection. IJCAI 14, 1137–1145 (1995)Google Scholar
  8. 8.
    Madnani, N., Tetreault, J., Chodorow, M.: Re-examining machine translation metrics for paraphrase identification. In: Proceedings of 2012 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2012), pp. 182–190 (2012)Google Scholar
  9. 9.
    Mohri, M., Rostamizadeh, A.: Stability bounds for non-i.i.d. processes. In: Advances in Neural Information Processing Systems, vol. 20, pp. 1025–1032 (2008)Google Scholar
  10. 10.
    Ng, A.Y.: Preventing overfitting of crossvalidation data. In: Proceedings of the 14th International Conference on Machine Learning, pp. 245–253 (1997)Google Scholar
  11. 11.
    Papineni, K., Roukos, S., Ward, T., Zhu, W.J.: BLEU: a method for automatic evaluation of machine translation. In: Proceedings of the 40th Annual meeting of the Association for Computational Linguistics (ACL-2102), pp. 311–318 (2002)Google Scholar
  12. 12.
    Quinonero-Candela, J., Sugiyama, M., Schwaighofer, A., Lawrence, N.D. (eds.): Dataset Shift in Machine Learning. Neural Information Processing series. MIT Press (2008)Google Scholar
  13. 13.
    Rao, R.B., Fung, G.: On the dangers of cross-validation. An experimental evaluation. In: Proceedings of the SIAM Conference on Data Mining, pp. 588–596 (2008)Google Scholar
  14. 14.
    Samuelson, F.: Supplementing a validation test sample. In: 2009 Joint Statistical Meetings (2009)Google Scholar
  15. 15.
    Stone, M.: Asymptotics for and against crossvalidation. Biometrika 64, 29–35 (1977)CrossRefMathSciNetzbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.IBM ResearchBangaloreIndia

Personalised recommendations