Encyclopedia of Machine Learning

2010 Edition
| Editors: Claude Sammut, Geoffrey I. Webb


  • Nicolas Lachiche
Reference work entry
DOI: https://doi.org/10.1007/978-0-387-30164-8_680


Propositionalization is the process of explicitly transforming a  Relational dataset into a propositional dataset.

The input data consists of examples represented by structured terms (cf.  Learning from Structured Data), several predicates in  First-Order Logic, or several tables in a relational database. We jointly refer to these as relational representations. The output is an  Attribute-value representation in a single table, where each example corresponds to one row and is described by its values for a fixed set of attributes. New attributes are often called features to emphasize that they are built from the original attributes. The aim of propositionalization is to pre-process relational data for subsequent analysis by attribute-value learners. There are several reasons for doing this, the most important of which are: to reduce the complexity and speed up the learning; to separate modeling the data from hypothesis construction; or to use familiar attribute-value (or...

This is a preview of subscription content, log in to check access.

Recommended Reading

  1. Dietterich, T. G., Lathrop, R. H., & Lozano-Pérez, T. (1997). Solving the multiple-instance problem with axis-parallel rectangles. Artificial Intelligence, 89(1–2), 31–71.zbMATHCrossRefGoogle Scholar
  2. Džeroski, S., & Lavrač, N. (Ed.). (2001). Relational data mining. Berlin: Springer.zbMATHGoogle Scholar
  3. Flach, P., & Lachiche, N. (1999). 1BC: A first-order bayesian classifier. In S. Džeroski & P. Flach (Eds.), Proceedings of the ninth international workshop on inductive logic programming (ILP’99), Vol. 1634 of lecture notes in computer science (pp. 92–103). Berlin: Springer.Google Scholar
  4. Knobbe, A. J., de Haas, M., & Siebes, A. (2001). Propositionalization and aggregates. In Proceedings of the sixth European conference on principles of data mining and knowledge discovery, Vol. 2168 of lecture notes in artificial intelligence (pp. 277–288). Berlin: Springer.Google Scholar
  5. Kramer, S., Lavrač, N., & Flach, P. (2001). Propositionalization approaches to relational data mining. In S. Džeroski & N. Lavrač (Eds.), Relational data mining (Chap. 11, pp. 262–291). Berlin: Springer.Google Scholar
  6. Krogel, M.-A., Rawles, S., Železný, F., Flach, P. A., Lavrač, N., & Wrobel, S. (2003). Comparative evaluation of approaches to propositionalization. In T. Horváth & A. Yamamoto (Eds.), Proceedings of the thirteenth international conference on inductive logic programming, Vol. 2835 of lecture notes in artificial intelligence (pp. 197–214). Berlin: Springer.Google Scholar
  7. Lachiche, N. (2005). Good and bad practices in propositionalization. In S. Bandini & S. Manzoni (Eds.), Proceedings of advances in artificial intelligence, ninth congress of the Italian association for artificial intelligence (AI*IA’05), Vol. 3673 of lecture notes in computer science (pp. 50–61). Berlin: Springer.Google Scholar
  8. Perlich, C., & Provost, F. (2006). Distribution-based aggregation for relational learning with identifier attributes. Machine Learning, 62, 62–105.CrossRefGoogle Scholar
  9. Srinivasan, A., Muggleton, S., King, R. D., & Stenberg, M. (1996). Theories for mutagenicity: A study of first-order and feature based induction. Artificial Intelligence, 85(1–2), 277–299.CrossRefGoogle Scholar
  10. Tomečková, M., Rauch, J., & Berka, P. (2002). Stulong data from longitudinal study of atherosclerosis risk factors. In P. Berka (Ed.), Discovery challenge workshop notes. ECML/PKDD’02, Helsinki, Finland. http://lisp.vse.cz/challenge/ecmlpkdd2002/proceedings/Tomeckova.pdf

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Nicolas Lachiche
    • 1
  1. 1.University of StrasbourgStrasbourgFrance