Abstract
A decision tree is a comprehensible representation that has been widely used in many supervised machine learning domains. But decision trees have two notable problems – those of replication and fragmentation. One way of solving these problems is to introduce the notion of decision graphs – a generalization of the decision tree – which addresses the above problems by allowing for disjunctions, or joins. While various decision graph systems are available, all of these systems impose some forms of restriction on the proposed representations, often leading to either a new redundancy or the original redundancy not being removed. Tan and Dowe (2002) introduced an unrestricted representation called the decision graph with multi-way joins, which has improved representative power and is able to use training data with improved efficiency. In this paper, we resolve the problem of encoding internal repeated structures by introducing dynamic attributes in decision graphs. A refined search heuristic to infer these decision graphs with dynamic attributes using the Minimum Message Length (MML) principle (see Wallace and Boulton (1968), Wallace and Freeman (1987) and Wallace and Dowe (1999)) is also introduced. On both real-world and artificial data, and in terms of both “right”/“wrong” classification accuracy and logarithm of probability “bit-costing” predictive accuracy (for binary and multinomial target attributes), our enhanced multi-way join decision graph program with dynamic attributes improves our Tan and Dowe (2002) multi-way join decision graph program, which in turn significantly out-performs both C4.5 and C5.0. The resultant graphs from the new decision graph scheme are also more concise than both those from C4.5 and from C5.0. We also comment on logarithm of probability as a means of scoring (probabilistic) predictions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Blake, C.L., Merz, C.J.: UCI repository of machine learning databases (1998), http://www.ics.uci.edu/~mlearn/MLRepository.html
Dowe, D.L., Farr, G.E., Hurst, A.J., Lentin, K.L.: Information-theoretic football tipping. In: de Mestre, N. (ed.) Third Australian Conference on Mathematics and Computers in Sport, pp. 233–241. Bond University, Qld, Australia (1996), http://www.csse.monash.edu.au/~footy
Dowe, D.L., Krusel, N.: A decision tree model of bushfire activity. In (Technical report 93/190) Dept Computer Science, Monash University, Clayton, Vic. 3800, Australia (1993)
Good, I.J.: Rational Decisions. Journal of the Royal Statistical Society. Series B 14, 107–114 (1952)
Good, I.J.: Corroboration, Explanation, Evolving Probability, Simplicity, and a Sharpened Razor. British Journal of Philosophy of Science 19, 123–143 (1968)
Kohavi, R.: Bottom-up induction of oblivious read-once decision graphs: Strengths and limitations. In: National Conference on Artificial Intelligence, pp. 613–618 (1994)
Mansour, Y., McAllester, D.: Boosting using branching programs. In: Proc. 13th Annual Conference on Comput. Learning Theory (CoLT), pp. 220–224. Morgan Kaufmann, San Francisco (2000)
Mehta, M., Rissanen, J., Agrawal, R.: MDL-based Decision Tree Pruning. In: The First International Conference on Knowledge Discovery & Data Mining, pp. 216–221. AAAI Press, Menlo Park (1995)
Needham, S.L., Dowe, D.L.: Message length as an effective Ockham’s razor in decision tree induction. In: Proc. 8th International Workshop on Artificial Intelligence and Statistics (AI+STATS 2001), Key West, Florida, U.S.A, January 2001, pp. 253–260 (2001)
Neil, J.R.: MML discovery of Causal Models. PhD thesis, Monash University, Clayton 3800, Australia, Computer Science and Software Engineering (2001)
Oliveira, A.L., Sangiovanni-Vincentelli, A.L.: Using the minimum description length principle to infer reduced ordered decision graphs. Machine Learning 25(1), 23–50 (1996)
Oliver, J.J.: Decision Graphs - An Extension of Decision Trees. In: Proc. 4th International Workshop on Artif. Intelligence and Statistics, pp. 343–350 (1993)
Oliver, J.J., Dowe, D.L., Wallace, C.S.: Inferring Decision Graphs Using the Minimum Message Length Principle. In: Proceedings of the 5th Joint Conference on Artificial Intelligence, pp. 361–367. World Scientific, Singapore (1992)
Oliver, J.J., Wallace, C.S.: Inferring Decision Graphs. In: Workshop 8 International Joint Conference on AI (IJCAI), Sydney, Australia (August 1991)
Perlich, C., Provost, F., Simonoff, J.S.: Tree induction versus logistic regression: a learning-curve analysis. Journal of Machine Learning Research 4, 211–255 (2003)
Provost, F., Domingos, P.: Tree induction for probability-based ranking. Machine Learning 52, 199–215 (2003)
Quinlan, J.R.: C4.5: Programs for Machine Learning. Morgan Kaufmann, San Mateo (1992), The latest version of C5 is available from, http://www.rulequest.com
Quinlan, J.R., Rivest, R.: Inferring Decision Trees Using the Minimum Description Length Principle. Information and Computation 80, 227–248 (1989)
Rissanen, J.J.: Modeling by shortest data description. Automatica 14, 465–471 (1978)
Tan, P.J., Dowe, D.L.: MML inference of decision graphs with multi-way joins. In: McKay, B., Slaney, J.K. (eds.) Canadian AI 2002. LNCS (LNAI), vol. 2557, pp. 131–142. Springer, Heidelberg (2002)
Uther, W.T.B., Veloso, M.M.: The Lumberjack Algorithm for Learning Linked Decision Tree. In: Mizoguchi, R., Slaney, J.K. (eds.) PRICAI 2000. LNCS (LNAI), vol. 1886, pp. 156–166. Springer, Heidelberg (2000)
Wallace, C.S., Boulton, D.M.: An Information Measure for Classification. Computer Journal 11, 185–194 (1968)
Wallace, C.S., Boulton, D.M.: An Invariant Bayes Method for Point Estimation. Classification Society Bull 3, 11–34 (1975)
Wallace, C.S., Dowe, D.L.: Minimum Message Length and Kolmogorov Complexity. Computer Journal, Special Issue - Kolmogorov Complexity 42(4), 270–283 (1999)
Wallace, C.S., Dowe, D.L.: MML Clustering of multi-state, Poisson, von Mises circular and Gaussian distributions. Statistics and Computing 10(1), 73–83 (2000)
Wallace, C.S., Freeman, P.R.: Estimation and Inference by Compact Coding. Journal of the Royal Statistical Society. Series B 49(3), 240–265 (1987)
Wallace, C.S., Patrick, J.D.: Coding Decision Trees. Machine Learning 11, 7–22 (1993)
Yin, X., Han, J.: CPAR: Classification based on predictive association rules. In: SIAM International Conference on Data Mining, San Francisco, CA, USA (May 2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Tan, P.J., Dowe, D.L. (2003). MML Inference of Decision Graphs with Multi-way Joins and Dynamic Attributes. In: Gedeon, T.(.D., Fung, L.C.C. (eds) AI 2003: Advances in Artificial Intelligence. AI 2003. Lecture Notes in Computer Science(), vol 2903. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-24581-0_23
Download citation
DOI: https://doi.org/10.1007/978-3-540-24581-0_23
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20646-0
Online ISBN: 978-3-540-24581-0
eBook Packages: Springer Book Archive