Abstract
Most data mining algorithms assume that their input data are described by a fixed number of attributes and each attribute has a pre-defined domain of values. However, the latter assumption is often not realistic in the case of categorical attributes. Such attributes are often available in different levels of granurality. The coarsest level might just have two possible values that are split into more values in refined levels of granularity. Before applying a data mining algorithm, it is usually assumed that the domain expert for the data must choose for each attribute the appropriate level of granularity or that in tedious trial and error procedure the appropriate granularity levels are adapted. The problem of choosing suitable granularity levels is related, but not identical to feature selection, since the more refined granularity levels increase the risk of overfitting. In this paper, we propose methods for decision and regression trees to handle the problem of different granularity levels during the construction of the corresponding tree.
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.
Similar content being viewed by others
References
Berthold, M.R., Borgelt, C., Höppner, F., Klawonn, F.: Guide to Intelligent Data Analysis: How to Intelligently Make Sense of Real Data. Springer, London (2010)
Breiman, L., Friedman, J., Olshen, R., Stone, C.: Classification and Regression Trees. Wadsworth, Belmont (1984)
Chaudhuri, S., Dayal, U.: An Overview of Data Warehousing and OLAP Technology. SIGMOD Rec. 28, 65–74 (1997)
Grünwald, P.D.: The Minimum Description Length Principle. MIT Press, Cambridge (2007)
Pedrycz, W., Skowron, A., Kreinovich, V. (eds.): Handbook of Granular Computing. Wiley, Chichester (2008)
Quinlan, J.R.: Induction of Decision Trees. Springer, New York (1986)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag GmbH Berlin Heidelberg
About this chapter
Cite this chapter
Ince, K., Klawonn, F. (2013). Decision and Regression Trees in the Context of Attributes with Different Granularity Levels. In: Borgelt, C., Gil, M., Sousa, J., Verleysen, M. (eds) Towards Advanced Data Analysis by Combining Soft Computing and Statistics. Studies in Fuzziness and Soft Computing, vol 285. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-30278-7_26
Download citation
DOI: https://doi.org/10.1007/978-3-642-30278-7_26
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-30277-0
Online ISBN: 978-3-642-30278-7
eBook Packages: EngineeringEngineering (R0)