Abstract
In this paper we present a method for induction of fuzzy characteristic rules from a fuzzy subset. We use a two step method where in the first step we search for candidate predicates which will make up the characteristic rules. For this purpose typical values are used to isolate characteristic intervals for attributes, these intervals will make up the candidate predicates. In the second step we use the candidate predicates to induce the fuzzy characteristic rules.
Chapter PDF
References
Han, J.: Mining Knowledge at Multiple Concept Levels. CIKM'94, Baltimore, Maryland, USA (1995) 19–24
Kamber, M. and Shinghal, R.: Evaluating the Interestingness of Characteristic Rules. Second Internationale Conference on Knowledge Discovery & Data Mining, Portland, Oregon (1996) 263–266
Han, J. and Fu, Y.: Attribute-Oriented Induction in Data Mining. Advances in Knowledge Discovery, AAAI Press/The MIT Press (1996) 399–421
Rasmussen, D. and Yager, R. R.: Summary SQL — A fuzzy tool for data mining. Intelligent Data Analysis (an electronic journal), (1996)
Michalski, R. S.: A Theory and Methodology of Inductive Learning. Artificial Intelligence 20 (1983) 111–161
Yager, R. R.: A note on a fuzzy measure of typicality. International Journal of Intelligent Systems 12 (1997) 233–249
Rasmussen, D. and Yager, R. R.: SummarySQL-A flexible fuzzy query language. FQAS'96, Roskilde, Denmark (1996) 1–18
Zadeh, L. A.: A theory of approximate reasoning. Machine Intelligence, Vol. 9, Hayes, J., Michie, D., & Mikulich, L.I. (eds.), New York: Halstead Press (1979) 149–194
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Rasmussen, D., Yager, R.R. (1997). Induction of fuzzy characteristic rules. In: Komorowski, J., Zytkow, J. (eds) Principles of Data Mining and Knowledge Discovery. PKDD 1997. Lecture Notes in Computer Science, vol 1263. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63223-9_112
Download citation
DOI: https://doi.org/10.1007/3-540-63223-9_112
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63223-8
Online ISBN: 978-3-540-69236-2
eBook Packages: Springer Book Archive