ecms_neu_mini.png

Digital Library

of the European Council for Modelling and Simulation

 

Title:

Classification Confidence Of Fuzzy Rule-Based Classifiers

Authors:

Tomoharu Nakashima, Ashish Ghosh

Published in:

 

(2011).ECMS 2011 Proceedings edited by: T. Burczynski, J. Kolodziej, A. Byrski, M. Carvalho. European Council for Modeling and Simulation. doi:10.7148/2011 

 

ISBN: 978-0-9564944-2-9

 

25th European Conference on Modelling and Simulation,

Jubilee Conference

Krakow, June 7-10, 2011

 

Citation format:

Nakashima, T., & Ghosh, A. (2011). Classification Confidence Of Fuzzy Rule-Based Classifiers. ECMS 2011 Proceedings edited by: T. Burczynski, J. Kolodziej, A. Byrski, M. Carvalho (pp. 466-471). European Council for Modeling and Simulation. doi:10.7148/2011-0466-0471

DOI:

http://dx.doi.org/10.7148/2011-0466-0471

Abstract:

In this paper we first introduce the concept of classifica- tion confidence in fuzzy rule-based classification. Clas- sification confidence shows the strength of classification for an unseen pattern. Low classification confidence for an unseen pattern means that the classification of that pat- tern is not very clear compared to that with high clas- sification confidence. Then we focus on the minimum classification confidence for fuzzy rule-based classifiers using the classification confidence. The minimum clas- sification confidence represents the worst classification among given training patterns. Some discussion on as- signing a weight to training pattern is given to show that cost-sensitive fuzzy rule-based classifiers are advanta- geous for producing a large minimum-confidence clas- sifiers. A series of experiments are done in order to show that reasonable classification boundaries can be obtained by cost-sensitive fuzzy rule-based classifiers if appropri- ate weights are assigned to training patterns.

Full text: