A fuzzy rule-based approach to the analysis of decision process and its application
Erik Chowdhury, Lonnie C. Ludeman · 1995
The basic problem addressed is that of developing an algorithm for a general purpose pattern recognition decision making system with a classifier that derives from a set of fuzzy linguistic rules based on expert knowledge and/or experience. In order to translate the fuzzy linguistic rules for the design of a fuzzy classifier, various ways for modeling the basic sentence connectives and, or and not and the conditional if... then... are investigated within the context of fuzzy set theory. The general purpose pattern recognition decision making system consists of five major units: pre-processor, fuzzifier, rule base, computation unit and decision maker. Each of these major blocks are discussed in detail. Finally, the experimental results concerning the problem of computer-assisted electrocardiogram (ECG) interpretation for positive identification of cardiac arrhythmias and the problem of computer vision system for identity verification are given that demonstrate the operation and performance of the proposed general purpose pattern recognition decision making system.