Type-2 Fuzzy Classifier Ensembles for Text Entailment

DOI

10.2991/jcis.2008.11How to use a DOI?

Keywords

type-2 fuzzy sets, classifier ensembles, fuzzy c-classification.

Abstract

This paper presents a new Type-2 Fuzzy Classifier ensemble, which enables to model parameter uncertainties by charac-terizing the fuzzy sets with secondary membership values. We use fuzzy clus-tering method to characterize primary membership values and genetic algorithm to approximate secondary membership grades. Furthermore, a weighing algo-rithm is used for a non-complex reduction for reasoning. We use transductive rea-soning, instead of inductive reasoning, to develop a local model for every new vec-tor, based on a nearness criterion vectors from the given database. It is shown that the method can improve classifier system modeling performance in comparison to well-known methods.

Copyright

© 2008, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Asli Celikyilmaz
AU  - I. Burhan Turksen
PY  - 2008/12
DA  - 2008/12
TI  - Type-2 Fuzzy Classifier Ensembles for Text Entailment
BT  - Proceedings of the 11th Joint Conference on Information Sciences (JCIS 2008)
PB  - Atlantis Press
SP  - 66
EP  - 72
SN  - 1951-6851
UR  - https://doi.org/10.2991/jcis.2008.11
DO  - 10.2991/jcis.2008.11
ID  - Celikyilmaz2008/12
ER  -