This paper explores the idea of closed world specialisation (CWS). While traditional CWS is performed as a postprocessing step, we propose two different approaches to incorporating it into the induction process of a bottom-up inductive logic programming system. The motivation comes from the fact that using CWS as a postprocessing step is incapable of solving problems in which the negated part of the hypothesis is crucial. We apply the proposed approaches to the ProGolem bottom-up ILP system. We give examples of problems, where classical CWS fails to find a complete and consistent solution, whereas the proposed approaches succeed. Tests on real-world datasets show that the proposed approaches perform at least as well as regular CWS, while being better in terms of predictive accuracy in some cases. We also point out some weaknesses of different CWS approaches.
AriasM., KhardonR. and MalobertiJ., Learning horn expressions with logan-h, Journal of Machine Learning Research8 (2007), 549-587.
2.
BainM. and MuggletonS., Non-monotonic learning, in: Inductive Logic Programming, MuggletonS., ed., Academic Press, 1992, pp. 145-161.
3.
BainM. and MuggletonS., Learning optimal chess strategies, in: Machine Intelligence 13, 1994, pp. 291-309.
4.
BergadanoF., GunettiD., NicosiaM. and RuffoG., Learning logic programs with negation as failure, in: Proceedings of the 5th International Workshop on Inductive Logic Programming, de RaedtL., ed., Department of Computer Science, Katholieke Universiteit Leuven, 1995, pp. 33-52.
5.
DemšarJ., Statistical comparisons of classifiers over multiple data sets, Journal of Machine Learning Research7 (2006), 1-30.
6.
InoueK. and KudohY., Learning extended logic programs, in: IJCAI, PollackM.E., ed., Morgan Kaufmann, 1997, pp. 176-181.
7.
LavračN. and DžeroskiS., Inductive Logic Programming - Techniques and Applications, Ellis Horwood series in artificial intelligence, Ellis Horwood, 1994.
8.
MuggletonS., Inverse entailment and progol, New Generation Computing13(3&4) (1995), 245-286.
9.
MuggletonS. and FengC., Efficient induction of logic programs, in: ALT, 1990, pp. 368-381.
10.
MuggletonS., SantosJ.C.A. and Tamaddoni-NezhadA., Progolem: A system based on relative minimal generalisation, in: ILP, RaedtL.D., ed., volume 5989 of Lecture Notes in Computer Science, Springer, 2009, pp. 131-148.
11.
PlotkinG., A further note on inductive generalization, in: Machine Intelligence, Vol. 6, Edinburgh University Press, 1971, pp. 101-124.
12.
QuinlanJ.R. and Cameron-JonesR.M., FOIL: A midterm report, in: ECML, BrazdilP.B., ed., volume 667 of Lecture Notes in Computer Science, Springer Berlin Heidelberg, 1993, pp. 3-20.
13.
RaedtL.D., Logical and Relational Learning, Cognitive Technologies, Springer, 2008.
14.
RaedtL.D. and BruynoogheM., On negation and three-valued logic in interactive concept-learning, in: Proceedings of the Ninth European Conference on Artificial Intelligence, AielloL., ed., Pitman, 1990, pp. 207-212.
15.
RaedtL.D. and LaerW.V., Inductive constraint logic, in: ALT, JantkeK.P., ShinoharaT. and ZeugmannT., eds, volume 997 of Lecture Notes in Computer Science, Springer, 1995, pp. 80-94.
SakamaC., Nonmonotonic inductive logic programming, in: LPNMR, EiterT., FaberW. and TruszczynskiM., eds, volume 2173 of Lecture Notes in Computer Science, Springer, 2001, pp. 62-80.
18.
SakamaC., Induction from answer sets in nonmonotonic logic programs, ACM Trans Comput Log6(2) (2005), 203-231.
19.
SakamaC. and InoueK., Brave induction: A logical framework for learning from incomplete information, Machine Learning76(1) (2009), 3-35.
20.
SeitzerJ., Stable ILP: Exploring the added expressivity of negation in the background knowledge, in: Proceedings of the Frontiers of Inductive Logic Programming Workshop at IJCAI'95, (1997).
21.
SrinivasanA., The Aleph Manual, 2004.
22.
SrinivasanA., MuggletonS. and BainM., Distinguishing exceptions from noise in non-monotonic learning, in: Proceedings of the 2nd International Workshop on Inductive Logic Programming, MuggletonS., ed., ICOT, 1992, pp. 97-107.
23.
Tamaddoni-NezhadA. and MuggletonS., The lattice structure and refinement operators for the hypothesis space bounded by a bottom clause, Machine Learning76(1) (2009), 37-72.
24.
TaylorK., Inverse resolution of normal clauses, in: Proceedings of the 3rd International Workshop on Inductive Logic Programming, MuggletonS., ed., Josef Stefan Institute, 1993, pp. 165-178.
