Activation of Fuzzy Rules in RETE Network
Zenon A. Sosnowski · Flexible Query Answering Systems · 2001
A central algorithm in rule-based expert systems is the pattern matching among rule predicates and current data. Systems like OPS5 and its various derivatives use the RETE algorithm for this function. This paper describes and analyses augmentations of the basic RETE algorithm that are incorporated into an experimental fuzzy expert system shell FuzzyCLIPS which contains the capabilities of handling fuzzy concepts and reasoning. The new concept of fuzzy RETE network allows all fuzzy rules, with the same linguistic variable, to be activated whenever fuzzy evidence concerning that variable is asserted into to the working memory.