Abstract
Several upgrades of Attribute-Value learning to Inductive Logic Programming have been proposed and used successfully. However, the Top-Down Data-Driven strategy, popularised by the AQ family, has not yet been transferred to ILP: if the idea of reducing the hypothesis space by covering a seed example is utilised with systems like PROGOL, Aleph or MIO, these systems do not benefit from the associated data-driven specialisation operator. This operator is given an incorrect hypothesis h and a covered negative example e and outputs a set of hypotheses more specific than h and correct wrt e. This refinement operator is very valuable considering heuristic search problems ILP systems may encounter when crossing plateaus in relational search spaces. In this paper, we present the data-driven strategy of AQ, in terms of a lgg-based change of representation of negative examples given a positive seed example, and show how it can be extended to ILP. We evaluate a basic implementation of AQ in the system Propal on a number of benchmark ILP datasets.
This is a preview of subscription content, log in via an institution to check access.
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
Alphonse, E.: Macro-opérateurs et Sélection Relationnelle en Programmation Logique Inductive: théorie et algorithmes. PhD thesis, Université Paris-Sud (2003)
Alphonse, E., Osmani, A.: On the connection between the phase transition of the covering test and the learning success rate. In: Proc. 16th Conf. of Inductive Logic Programming (2006)
Alphonse, E., Rouveirol, C.: Lazy propositionalization for relational learning. In: Proc. ECAI 2000, pp. 256–260. IOS Press, Amsterdam (2000)
Blockeel, H., De Raedt, L.: Top-down induction of first order decision trees. Artificial Intelligence 101, 285–297 (1998)
Clark, P., Niblett, T.: The CN2 induction algorithm. Machine Learning 3, 261–283 (1989)
Cohen, W.W.: Fast effective rule induction. In: Proc. 12th ICML, pp. 115–123. Morgan Kaufmann, San Francisco (1995)
de Givry, S., Larrosa, J., Meseguer, P., Schiex, T.: Solving Max-SAT as weighted CSP. In: Rossi, F. (ed.) CP 2003. LNCS, vol. 2833, pp. 363–376. Springer, Heidelberg (2003)
de Givry, S., Zytnicki, M., Heras, F., Larrosa, J.: Existential arc consistency: Getting closer to full arc consistency in weighted CSP. In: Proc. of IJCAI-05 (2005)
Eisinger, N.: Subsumption and connection graphs. In: Proc. of IJCA 1981, pp. 480–486. William Kaufmann (1981)
Esposito, F., Laterza, A., Malerba, D., Semeraro, G.: Refinement of Datalog programs. In: Proc. of the MLnet Familiarization Workshop on ILP for KDD, pp. 73–94 (1996)
Fürnkranz, J.: Pruning methods for rule learning algorithms. In: Proc. 4th Int. Workshop on ILP, pp. 321–336 (1994)
Geibel, P., Wysotzki, F.: A Logical Framework for Graph Theoretical Decision Tree Learning. In: Proc. ILP 1997 (1997)
Giordana, A., Saitta, L., Sebag, M., Botta, M.: Analyzing relational learning in the phase transition framework. In: Proc. ICML, pp. 311–318 (2000)
Haussler, D.: Learning conjunctive concepts in structural domains. Machine Learning 4(1), 7–40 (1989)
Jagota, A.: Constraint satisfaction and maximum clique. In: Working Notes, AAAI Spring Symposium on AI and NP-hard Problems, pp. 92–97 (1993)
Kaufman, K.A., Michalski, R.S.: Learning from inconsistent and noisy data: The AQ18 approach. In: Proc. of the Eleventh ISMIS, pp. 411–419 (1999)
Kietz, J.-U.: A comparative study of structural most specific generalisations used in machine learning. In: Proc. Third Workshop on ILP, pp. 149–164 (1993)
Kietz, J.-U.: Some computational lower bounds for the computational complexity of inductive logic programmming. In: Proc. 6th ECML, Vienna, Austria (1993)
Kramer, S., Lavrac, N., Flach, P.: Propositionalization approaches to relational data mining. In: Dzeroski, S., Lavrac, N. (eds.) Relational Data Mining, pp. 262–291. Springer, Heidelberg (2001)
Krogel, M.: On Propositionalization for Knowledge Discovery in Relational Databases. PhD thesis, Univ. Magdeburg (2005)
Lloyd, J.W.: Foundations of Logic Programming, 2nd edn. Springer, Berlin (1987)
Michalski, R.S.: A theory and methodology of inductive learning. Machine Learning: An Artificial Intelligence Approach I, 83–134 (1983)
Michalski, R.S., Wojtusiak, J.: Reasoning with meta-values in AQ learning. Technical report, George Mason University (2006)
Muggleton, S.: Inverse entailment and PROGOL. New Generation Computing 13, 245–286 (1995)
Muggleton, S.H., Bain, M., Hayes-Michie, J., Michie, D.: An experimental comparison of human and machine learning formalisms. In: Proc. 6th IWML, San Mateo, CA, pp. 113–118. Morgan Kaufmann, San Francisco (1989)
Castillo, L.P., Wrobel, S.: On the stability of example-driven learning systems: A case study in multirelational learning. In: Coello Coello, C.A., de Albornoz, Á., Sucar, L.E., Battistutti, O.C. (eds.) MICAI 2002. LNCS (LNAI), vol. 2313, pp. 321–330. Springer, Heidelberg (2002)
Plotkin, G.: A note on inductive generalization. In: Machine Intelligence, vol. 5, Edinburgh University Press, Edinburgh (1970)
Quinlan, J.R.: Learning logical definitions from relations. Machine Learning 5(3), 239–266 (1990)
Scheffer, T., Herbrich, R.: Unbiased assessment of learning algorithms. In: Proc. Int. Joint Conf. on Artificial Intelligence (IJCAI 1997), pp. 798–803 (1997)
Scheffer, T., Herbrich, R., Wysotzki, F.: Efficient θ-subsumption based on graph algorithms. In: Inductive Logic Programming. LNCS, vol. 1314, pp. 312–329. Springer, Heidelberg (1997)
Sebag, M.: Delaying the choice of bias: a disjunctive version space approach. In: Proc. 13th ICML, pp. 444–452 (1996)
Sebag, M., Rouveirol, C.: Constraint inductive logic programming. In: Advances In Inductive Logic Programming, pp. 277–294. IOS Press, Amsterdam (1996)
Sebag, M., Rouveirol, C.: Resource-bounded relational reasoning: Induction and deduction through stochastic matching. Machine Learning 38(1/2), 41–62 (2000)
Smith, B.D., Rosenbloom, P.S.: Incremental non-backtracking focusing: A polynomially bounded generalization algorithm for version spaces. In: Proc. 8th AAAI, pp. 848–853 (1990)
Srinivasan, A.: A learning engine for proposing hypotheses (Aleph) (1999)
Srinivasan, A., Muggleton, S., King, R.D.: Comparing the use of background knowledge by inductive logic programming systems. In: De Raedt, L. (ed.) Proc. of the 5th ILP Workshop, pp. 199–230. Scientific Report, K.U.Leuven (1995)
van der Laag, P.R.J., Nienhuys-Cheng, S-H.: Existence and nonexistence of complete refinement operators. In: Bergadano, F., De Raedt, L. (eds.) Machine Learning: ECML-94. LNCS, vol. 784, pp. 307–322. Springer, Heidelberg (1994)
VanLehn, K.: Efficient specialization of relational concepts. Machine Learning 4, 99–106 (1989)
Winston, P.H.: Learning structural descriptions form examples. In: Winston, P.H. (ed.) The Psychology of Computer Vision, pp. 157–209. McGraw-Hill, New York (1975)
Zucker, J.-D., Ganascia, J.-G.: Selective reformulation of examples in concept learning. In: Proc. 11th ICML, pp. 352–360 (1994)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Alphonse, E., Rouveirol, C. (2007). Extension of the Top-Down Data-Driven Strategy to ILP. In: Muggleton, S., Otero, R., Tamaddoni-Nezhad, A. (eds) Inductive Logic Programming. ILP 2006. Lecture Notes in Computer Science(), vol 4455. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-73847-3_13
Download citation
DOI: https://doi.org/10.1007/978-3-540-73847-3_13
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-73846-6
Online ISBN: 978-3-540-73847-3
eBook Packages: Computer ScienceComputer Science (R0)