Rule-Based Prediction of Rare Extreme Values
This paper describes a rule learning method that obtains models biased towards a particular class of regression tasks. These tasks have as main distinguishing feature the fact that the main goal is to be accurate at predicting rare extreme values of the continuous target variable. Many real-world applications from scientific areas like ecology, meteorology, finance,etc., share this objective. Most existing approaches to regression problems search for the model parameters that optimize a given average error estimator (e.g. mean squared error). This means that they are biased towards achieving a good performance on the most common cases. The motivation for our work is the claim that being accurate at a small set of rare cases requires different error metrics. Moreover, given the nature and relevance of this type of applications an interpretable model is usually of key importance to domain experts, as predicting these rare events is normally associated with costly decisions. Our proposed system (R-PREV) obtains a set of interpretable regression rules derived from a set of bagged regression trees using evaluation metrics that bias the resulting models to predict accurately rare extreme values. We provide an experimental evaluation of our method confirming the advantages of our proposal in terms of accuracy in predicting rare extreme values.
KeywordsRegression Tree Domain Expert Target Variable Minority Class Interpretable Model
Unable to display preview. Download preview PDF.
- 1.Breiman, L., Friedman, J., Olshen, R., Stone, C.: Classification and Regression Trees. Statistics/Probability Series. Wadsworth & Brooks/Cole Advanced Books & Software (1984)Google Scholar
- 2.Cleveland, W.: Visualizing data. Hobart Press (1993)Google Scholar
- 3.Demšar, J.: Statistical comparisons of classifiers over multiple data sets. Journal of Machine Learning Research 7, 1–30 (2006)Google Scholar
- 6.Ribeiro, R.: Prediction models for rare phenomena. Master’s thesis, Faculty of Economics, University of Porto, Portugal (February 2004)Google Scholar
- 7.Van Rijsbergen, C.: Information Retrieval, 2nd edn. Dept. of Computer Science, University of Glasgow (1979)Google Scholar
- 8.Torgo, L., Gama, J.: Regression using classification algorithms. Intelligent Data Analysis 1(4) (1997)Google Scholar
- 10.Turney, P.: Types of cost in inductive learning. In: Proceedings of the Workshop on cost-sensitive learning at the 17th ICML, pp. 15–21 (2000)Google Scholar
- 11.UCI Machine Learning Repository, http://www.ics.uci.edu/~mlearn/MLSummary.html
- 12.Weiss, G., Hirsh, H.: Learning to predict rare events in categorical time-series data. In: AAAI Workshop on Predicting the Future: AI Approaches to Time-Series Problems, vol. WS-98-07, pp. 83–90. AAAI Press, Menlo Park (1998)Google Scholar
- 13.Weiss, G., Hirsh, H.: Learning to predict extremely rare events. In: AAAI Workshop on Learning from Imbalanced Data Sets, vol. WS-00-05, pp. 64–68. AAAI Press, Menlo Park (2000)Google Scholar
- 14.Weiss, G., Provost, F.: The effect of class distribution on classifier learning: an empirical study. Technical Report ML-TR-44, Department of Computer Science, Rutgers University (2001)Google Scholar