Abstract
This paper presents the integration of complex aggregates in the construction of logical decision trees to address relational data mining tasks. Indeed, relational data mining implies aggregating properties of objects from secondary tables and complex aggregates are an expressive way to do so. However, the size of their search space is combinatorial and it cannot be explored exhaustively. This leads us to introduce a new algorithm to build relevant complex aggregate features. This algorithm uses random restart hill-climbing to build complex aggregation conditions. The algorithm shows good results on both artificial data and real-world data.
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 subscriptionsNotes
- 1.
This concept of branch has nothing to do with the concept of branch in a decision tree. We refer here to a part of the complex aggregate search space.
References
Assche, A.V., Vens, C., Blockeel, H., Dzeroski, S.: First order random forests: learning relational classifiers with complex aggregates. Mach. Learn. 64(1–3), 149–182 (2006)
Blockeel, H., Dehaspe, L., Ramon, J., Struyf, J., Assche, A.V., Vens, C., Fierens, D.: The ACE Data Mining System, March 2009
Blockeel, H., Raedt, L.D.: Top-down induction of first-order logical decision trees. Artif. Intell. 101(1–2), 285–297 (1998)
Getoor, L.: Multi-relational data mining using probabilistic relational models: research summary. In: Proceedings of the First Workshop in Multi-relational Data Mining (2001)
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The Weka data mining software: an update. SIGKDD Explor. Newslett. 11(1), 10–18 (2009). http://doi.acm.org/10.1145/1656274.1656278
El Jelali, S., Braud, A., Lachiche, N.: Propositionalisation of continuous attributes beyond simple aggregation. In: Riguzzi, F., Železný, F. (eds.) ILP 2012. LNCS, vol. 7842, pp. 32–44. Springer, Heidelberg (2013)
Joshi, S., Ramakrishnan, G., Srinivasan, A.: Feature construction using theory-guided sampling and randomised search. In: Železný, F., Lavrač, N. (eds.) ILP 2008. LNCS (LNAI), vol. 5194, pp. 140–157. Springer, Heidelberg (2008). http://dx.doi.org/10.1007/978-3-540-85928-4_14
Krogel, M.A., Wrobel, S.: Facets of aggregation approaches to propositionalization. In: Horvath, T., Yamamoto, A. (eds.) Work-in-Progress Track at the Thirteenth International Conference on Inductive Logic Programming (ILP) (2003)
Puissant, A., Lachiche, N., Skupinski, G., Braud, A., Perret, J., Mas, A.: Classification et évolution des tissus urbains à partir de données vectorielles. Revue Internationale de Géomatique 21(4), 513–532 (2011)
Ruas, A., Perret, J., Curie, F., Mas, A., Puissant, A., Skupinski, G., Badariotti, D., Weber, C., Gancarski, P., Lachiche, N., Lesbegueries, J., Braud, A.: Conception of a GIS-platform to simulate urban densification basedon the analysis of topographic data. In: Ruas, A. (ed.) Advances in Cartography and GIScience. LNGC, vol. 2, pp. 413–430. Springer, Heidelberg (2011). http://dx.doi.org/10.1007/978-3-642-19214-2_28
Vens, C., Ramon, J., Blockeel, H.: Refining aggregate conditions in relational learning. In: Fürnkranz, J., Scheffer, T., Spiliopoulou, M. (eds.) PKDD 2006. LNCS (LNAI), vol. 4213, pp. 383–394. Springer, Heidelberg (2006)
Acknowledgements
This work is part of the REFRAME project granted by the European Coordinated Research on Long-term Challenges in Information and Communication Sciences & Technologies ERA-Net (CHIST-ERA).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Charnay, C., Lachiche, N., Braud, A. (2015). Construction of Complex Aggregates with Random Restart Hill-Climbing. In: Davis, J., Ramon, J. (eds) Inductive Logic Programming. Lecture Notes in Computer Science(), vol 9046. Springer, Cham. https://doi.org/10.1007/978-3-319-23708-4_4
Download citation
DOI: https://doi.org/10.1007/978-3-319-23708-4_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-23707-7
Online ISBN: 978-3-319-23708-4
eBook Packages: Computer ScienceComputer Science (R0)