Date of Award


Degree Name

Master of Science


Computer Science

First Advisor

Rahimi, Shahram


Computing with words, or CW, is a method of using words derived from a natural language [1] rather than numbers for computing and reasoning, CW simulates the way in which humans express and use their knowledge since its processing is close to that of a natural language, using a mathematical model to produce meanings for inexact words and phrases. Computing with words has been used in numerous applications such as decision analysis. In a CW Inference Engine, since the inputs are linguistic words, the outputs should be linguistic expressions so as to make more sense to humans [2]. In this thesis, I have adjusted and implemented the method introduced in [3] for generating linguistic expressions for fuzzy output sets produced by CW inference. The process for finding linguistic labels is called linguistic approximation (LA). Linguistic approximation plays an essential role in finding the best output fuzzy label. It provides a solution for the problem of finding the meaning of a given fuzzy set. In general, it is required to map a given fuzzy value (which is an output of some fuzzy computation) to a set of predefined terms, defined on the same universe of discourse [9]. The result of a linguistic approximation algorithm will be linguistic expressions, atomic terms and linguistic modifiers [3]. The algorithm works by dividing the membership function into segments and then finding the most appropriate linguistic expressions corresponding to the meaning of the fuzzy set.




This thesis is only available for download to the SIUC community. Others should
contact the interlibrary loan department of your local library.