TY  - CONF
ID  - moewes_usefulness_2011
T1  - On the usefulness of fuzzy SVMs and the extraction of fuzzy rules from SVMs
A1  - Moewes, Christian
A1  - Kruse, Rudolf
ED  - Galichet, Sylvie
ED  - Montero, Javier
ED  - Mauris, Gilles
TI  - Proceedings of the 7th conference of the European Society for Fuzzy Logic and Technology (EUSFLAT-2011) and LFA-2011
T3  - Advances in Intelligent Systems Research
Y1  - 2011
VL  - 17
SP  - 943
EP  - 948
PB  - Atlantis Press
T2  - European Society for Fuzzy Logic and Technology (EUSFLAT)
CY  - Aix-les-Bains, France
AD  - Amsterdam / Paris
SN  - 978-90-78677-00-0
SN  - 1951-6851
UR  - http://www.atlantis-press.com/publications/aisr/eusflat-11/
M2  - doi: doi:10.2991/eusflat.2011.46
KW  - Binary Classification
KW  - Classification
KW  - Fuzzy Rules
KW  - fuzzy SVM
KW  - Support Vector Machine
N2  - In this paper we reason about the usefulness of two recent trends in fuzzy methods in machine learning. That is, we discuss both fuzzy support vector machines (FSVMs) and the extraction of fuzzy rules from SVMs. First, we show that an FSVM is identical to a special type of SVM. Second, we categorize and analyze existing approaches to obtain fuzzy rules from SVMs. Finally, we question both trends and conclude with more promising alternatives.
ER  -