Generating Rule Sets from Model Trees
Model trees—decision trees with linear models at the leaf nodes—have recently emerged as an accurate method for numeric prediction that produces understandable models. However, it is known that decision lists—ordered sets of If-Then rules—have the potential to be more compact and therefore more understandable than their tree counterparts.
We present an algorithm for inducing simple, accurate decision lists from model trees. Model trees are built repeatedly and the best rule is selected at each iteration. This method produces rule sets that are as accurate but smaller than the model tree constructed from the entire dataset. Experimental results for various heuristics which attempt to find a compromise between rule accuracy and rule coverage are reported. We show that our method produces comparably accurate and smaller rule sets than the commercial state-of-the-art rule learning system Cubist.
KeywordsModel Tree Multivariate Adaptive Regression Spline Projection Pursuit Percent Root Decision List
Unable to display preview. Download preview PDF.
- 2.W.W. Cohen. Fast effective rule induction. In Proc. of the Twelfth International Conference on Machine Learning, pages 115–123. Morgan Kaufmann, 1995.Google Scholar
- 3.G. Das, K. I. Lin, G. Renganathan, and P. Smyth. Rule discovery from time series. In Proc. of the Fourth International Conference on Knowledge Discovery and Data Mining, pages 16–22. AAAI Press, 1998.Google Scholar
- 4.E. Frank and I. H. Witten. Generating accurate rule sets without global optimization. In Proc. of the Fifteenth International Conference on Machine Learning, pages 144–151. Morgan Kaufmann, 1998.Google Scholar
- 5.J. Freidman. Multivariate adaptive regression splines. Annals of Statistics, 19(1):1–141, 1991.Google Scholar
- 7.A. Karalic. Employing linear regression in regression tree leaves. In Proc. of the Tenth European Conference on Artificial Intelligence, Vienna, Austria, 1992.Google Scholar
- 8.E.J. Keogh and M. J. Pazzani. An enhanced representation of time series which allows fast and accurate classification, clustering and relevance feedback. In Proc. of the Fourth International Conference on Knowledge Discovery and Data Mining, pages 239–243. AAAI Press, 1998.Google Scholar
- 9.D. Kilpatrick and M. Cameron-Jones. Numeric prediction using instance-based learning with encoding length selection. In Nikola Kasabov, Robert Kozma, Kitty Ko, Robert O’Shea, George Coghill, and Tom Gedeon, editors, Progress in Connectionist-Based Information Systems, volume 2, pages 984–987. Springer-Verlag, 1998.Google Scholar
- 10.J. R. Quinlan. Learning with continuous classes. In Proc. of the Fifth Australian Joint Conference on Artificial Intelligence, pages 343–348, World Scientific, Singapore, 1992.Google Scholar
- 11.J. R. Quinlan. C4.5: Programs for Machine Learning. Morgan Kaufmann, San Mateo, CA., 1993.Google Scholar
- 13.StatLib. Department of Statistics, Carnegie Mellon University, 1999. http://lib.stat.cmu.edu.
- 14.L. Torgo. Data fitting with rule-based regression. In J. Zizka and P. Brazdil, editors, Proc. of the Workshop on Artificial Intelligence Techniques (AIT’95), Brno, Czech Republic, 1995.Google Scholar
- 15.Y. Wang and I. H. Witten. Induction of model trees for predicting continuous classes. In Proc. of the poster papers of the European Conference on Machine Learning, pages 128–137, Prague, Czech Republic, 1997.Google Scholar