HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Conference papers

A Study of Heuristic Evaluation Measures in Fuzzy Rule Induction

Abstract : The rule induction process could be conceived as a search process, and hence an evaluation metric is needed to estimate the quality of rules found in the search space and to direct the search towards the best rule. The evaluation measure is the most influential inductive bias in rule learning. It is therefore important to investigate its influence on the induction process and to compare the behaviour of different evaluation measures. Many different evaluation measures have been used to score crisp rules. For some of these measures, fuzzy variations have been designed and used to score fuzzy rules. This paper examines the most popular crisp evaluation measures and demonstrates how they can be adapted into the fuzzy domain. The paper also studies the performance of these measures on a large number of data sets when used in a recently developed fuzzy rule induction algorithm. Results show that there are no universally applicable evaluation measures and the choice of the best measure depends on the type of the data set and the learning problem.
Document type :
Conference papers
Complete list of metadata

Cited literature [20 references]  Display  Hide  Download

Contributor : Hal Ifip Connect in order to contact the contributor
Submitted on : Friday, June 22, 2018 - 11:46:10 AM
Last modification on : Friday, June 22, 2018 - 12:00:38 PM
Long-term archiving on: : Monday, September 24, 2018 - 8:16:49 PM


Files produced by the author(s)


Distributed under a Creative Commons Attribution 4.0 International License



Ashraf Afifi. A Study of Heuristic Evaluation Measures in Fuzzy Rule Induction. 14th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), May 2018, Rhodes, Greece. pp.533-545, ⟨10.1007/978-3-319-92007-8_45⟩. ⟨hal-01821074⟩



Record views


Files downloads