Abstract
(Multi-)relational regression consists of predicting continuous response of target objects called reference objects by taking into account interactions with other objects called task-relevant objects. In relational databases, reference objects and task-relevant objects are stored in distinct data relations. Interactions between objects are expressed by means of (many-to-one) foreign key constraints which may allow linking explanatory variables of a task-relevant object in several alternative ways to the response variable. By materializing multiple assignments in distinct attribute-value vectors, a reference object is represented as a bag of multiple instances, although there is only one response value for the entire bag. This works points out the same assumption of multi-instance learning that is a primary instance is responsible for the observed response value of a reference object. We propose a top-down induction multi-relational model tree system which navigates foreign key constraints according to a divide-and-conquer strategy, derives a representation of reference objects as bags of attribute-value vectors and then, for each bag, constructs a primary instance as main responsible of the response value. Coefficients of local hyperplane are estimated in an EM implementation of the stepwise least square regression. Experiments confirm the improved accuracy of our proposal with respect to traditional attribute-value and relational model tree learners.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Appice, A.: Learning Relational model Trees. PhD thesis, Department of Computer Science, University of Bari, Bari, Italy (2005)
Blockeel, H.: Top-down induction of first order logical decision trees. PhD thesis, Department of Computer Science, Katholieke Universiteit, Leuven, Belgium (1998)
Blockeel, H., Page, D., Srinivasan, A.: Multi-instance tree learning. In: Raedt, L.D., Wrobel, S. (eds.) International Conference on Machine Learning, vol. 119, pp. 57–64. ACM, New York (2005)
Breiman, L., Friedman, J., Olshen, R., Stone, J.: Classification and regression tree. Wadsworth & Brooks (1984)
Davis, J., Costa, V.S., Ray, S., Page, D.: An integrated approach to feature invention and model construction for drug activity prediction. In: Ghahramani, Z. (ed.) International Conference on Machine Learning, vol. 227, pp. 217–224. ACM, New York (2007)
Dietterich, T.G., Lathrop, R.H., Lozano-Pérez, T.: Solving the multiple instance problem with axis-parallel rectangles. Artif. Intell. 89(1-2), 31–71 (1997)
Dooly, D.R., Zhang, Q., Goldman, S.A., Amar, R.A.: Multiple-instance learning of real-valued data. Journal of Machine Learning Research 3, 651–678 (2002)
Draper, N.R., Smith, H.: Applied regression analysis. John Wiley & Sons, Chichester (1982)
Driessens, K., Džeroski, S.: Combining model-based and instance-based learning for first order regression. In: Raedt, L.D., Wrobel, S. (eds.) International Conference on Machine Learning, ICML 2005, pp. 193–205. ACM, New York (2005)
Džeroski, S., Blockeel, H., Kramer, S., Kompare, B., Pfahringer, B., Van Laer, W.: Experiments in predicting biodegradability. In: Džeroski, S., Flach, P.A. (eds.) ILP 1999. LNCS (LNAI), vol. 1634, pp. 80–91. Springer, Heidelberg (1999)
Džeroski, S., Lavrač, N.: Relational Data Mining. Springer, Heidelberg (2001)
Džeroski, S., Todoroski, L., Urbancic, T.: Handling real numbers in inductive logic programming: A step towards better behavioural clones. In: Lavrač, N., Wrobel, S. (eds.) ECML 1995. LNCS, vol. 912, pp. 283–286. Springer, Heidelberg (1995)
Hardle, W.: Applied nonparametric Regression. Cambridge University Press, Cambridge (1990)
Kramer, S.: Relational Learning vs. Propositionalization: Investigations in Inductive Logic Programming and Propositional Machine Learning. PhD thesis, Vienna University of Technology, Vienna, Austria (1999)
Kramer, S., Pfahringer, B., Helma, C.: Stochastic propositionalization of non-determinate background knowledge. In: Page, D.L. (ed.) ILP 1998. LNCS, vol. 1446, pp. 80–94. Springer, Heidelberg (1998)
Maron, O., Lozano-Perez, T.: A framework for multi-instance learning. Advanced in Neural Information Processing Systems
Orkin, M., Drogin, R.: Vital Statistics. McGraw-Hill, New York (1990)
Ray, S., Page, D.: Multiple instance regression. In: Brodley, C.E., Danyluk, A.P. (eds.) International Conference on Machine Learning, pp. 425–432. Morgan Kaufmann, San Francisco (2001)
Saith, R., Sergent, I.: Embryo selection for transfer in human IVF. Assist Reprod. Rev. 5, 145–154 (1995)
Srinivasan, A., Camacho, R.: Numerical reasoning with an ilp system capable of lazy evaluation and customised search. J. Log. Program 40(2-3), 185–213 (1999)
Srinivasan, A., Muggleton, S., King, R.D., Sternberg, M.J.E.: Mutagenesis: ILP experiments in a non-determinate biological domain. In: Wrobel, S. (ed.) International Workshop on Inductive Logic Programming, pp. 217–232 (1994)
Vens, C., Ramon, J., Blockeel, H.: Remauve: A relational model tree learner. In: Muggleton, S., Otero, R., Tamaddoni-Nezhad, A. (eds.) ILP 2006. LNCS (LNAI), vol. 4455, pp. 424–438. Springer, Heidelberg (2007)
Zhang, Q., Goldman, S.: EM-DD: An improved multiple-instance learning technique. In: Neural Information Processing Systems (2001)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Appice, A., Ceci, M., Malerba, D. (2008). Top-Down Induction of Relational Model Trees in Multi-instance Learning. In: Železný, F., Lavrač, N. (eds) Inductive Logic Programming. ILP 2008. Lecture Notes in Computer Science(), vol 5194. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-85928-4_7
Download citation
DOI: https://doi.org/10.1007/978-3-540-85928-4_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-85927-7
Online ISBN: 978-3-540-85928-4
eBook Packages: Computer ScienceComputer Science (R0)