Statistical-Like Validation of Fuzzy Logic Models Using Distributional Fitting

Armin Shmilovici, Haim Shore, Oded Maimon

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A statistical-like validation procedure is developed for fuzzy logic models. The procedure is based on approximating the unknown probability distribution of the model's residual errors, and the similarity measures of the errors, using a new distributional fitting procedure. The fuzzy logic models are validated using critical ratios generated from the fitted distribution.
Original languageEnglish GB
Title of host publicationInformation Processing and Management of Uncertainty in Knowledge-based Systems (IPMU'96)
Pages629-634
VolumeProceedings
StatePublished - 4 Jul 1996

Fingerprint

Dive into the research topics of 'Statistical-Like Validation of Fuzzy Logic Models Using Distributional Fitting'. Together they form a unique fingerprint.

Cite this