Solving Selection Problems Using Preference
Relation Based on Bayesian Learning

Tomofumi Nakano and Nobuhiro Inuzuka

Nagoya Institute of Technology,
Gokiso-cho, Showa-ku, Nagoya 466-8555, Japan
{tnakano, inuzuka}@ics.nitech.ac.jp



Abstract. This paper defines a selection problem which selects an 
appropriate object from a set that is specified by parameters. We discuss
inductive learning of selection problems and give a method combining
inductive logic programming (ILP) and Bayesian learning. It induces
a binary relation comparing likelihood of objects being selected. Our
methods estimate probability of each choice by evaluating variance of an
induced relation from an ideal binary relation. Bayesian learning 
combines a prior probability of objects and the estimated probability. By
making several assumptions on probability estimation, we give several
methods. The methods are applied to Part-of-Speech tagging.
References

1.	J. Cussens. Part-of-Speech Disambiguation Using ILP, Technical Report PRG-
TR-25-96, Oxford University Computing Laboratory, 1996.
2.	J. Cussens. Bayesian Inductive Logic Programming with Explicit Probabilistic
Bias, Technical Report PRG-TR-24-96, Oxford University Computing Laboratory, 1996.
3.	J. Cussens. Part-of-Speech Tagging Using Progol, Proc. 7th Intl Workshop on
ILP, LNAI 1297, pp. 93108, Springer, 1997
4.	J. Cussens. Using Prior Probabilities and Density Estimation for Relational 
Classification, Proc. 8th Intl Conf. on ILP, LNAI 1446, Springer, pp. 106115, 1998.
5.	J. Cussens, S. Dzeroski and T. Erjavec. Morphosintactic Tagging of Solovene
Using Progol, Proc. 9th Intl Workshop on ILP, LNAI 1634, Springer, pp. 6879,
1999.
6.	H. A. David. Ranking from Unbalanced Paired-Comparison Data, Biometrika
74, pp. 432436, 1987
7.	L. Dehaspe. Maximum Entropy Modeling with Clausal Constraints, Proc. 7th
Intl Workshop on ILP, LNAI 1297, Springer, pp. 109124, 1997.
8.	P. Flach and N. Lachiche. 1BC: A First Order Bayesian Classifier, Proc. 9th
ILP, LNAI 1634, Springer, pp. 92103, 1999.
9.	J. D. Hirst, R. D. King and M. J. B. Sternberg, Quantitative structure-activity
relationships by neural networks and inductive logic programming. I. The inhibition
of dihydrofolate reductase by pyrimidines, Journal of Computer-Aided Molecular
Design 8 (1994), pp. 405-420, 1994.
10.	T. Hotvth, Z. Alexin, T. Gyim6thy and S. Wrobel. Application of Different
Learning Methods to Hungarian Part-of-Speech Tagging, Proc. 9th Intl 
Workshop on ILP, LNAI 1634, Springer, pp. 128139, 1999.
11.	N. Inuzuka, T. Nakano and H. Itoh. Acquiring Heuristic Value in Search Trees by a
Relational Learner, Proc. 11th Intl Conf. on Applications of Prolog, pp. 157-162,
1998.
12.	J Jelonek and J Stefanowski. Experiments on Solving Multiclass Learning 
Problems by n2-classifier, Proc. 10th European Conf. on Machine Learning, LNAI
1398, Springer, pp. 172177, 1998.
13.	D. Koller. Probabilistic Relational Models . Proc. 9th ILP, LNAI 1634, Springer,
pp. 313, 1999.
14.	L. Mrquetz and H. Rodriguez. Part-of-Speech Tagging Using Decision Trees,
Proc. 10th European Conf. on Machine Learning, LNAI 1398, Springer, pp. 2536,
1998.
15.	S. Muggleton. A Learning Model for Universal Representations, Proc. 4th Intl
Workshop on ILP, pp. 139160, 1994.
16.	S. Muggleton. Inverse Entailment and Progol, New Generation Computing 
Journal, 13 pp. 245286, 1995.
17.	T. Nakano, N. Inuzuka, H. Seki and H. Itoh. Inducing Shogi Heuristics Using
Inductive Logic Programming, Proc. 8th Intl Conf. on ILP, LNAI 1446, Springer,
pp. 155164. 1998.
18.	S. Slattery and M Craven. Combining Statistical and Relational Methods for
Learning in Hypertext Domains, Proc. 8th Intl Conf. on ILP, LNAI 1446,
Springer, pp. 3852, 1998.
