Sidebar: Seven Truths of Fuzzy Logic

1. Fuzzy logic isn't fuzzy. Fuzzy logic isn't intrinsically imprecise, doesn't violate common sense and produces unambiguous results. "Classical" Boolean logic, in fact, is merely a special case of fuzzy logic.

2. Fuzzy logic is different from probability. With probability, we're trying to determine something about the potential outcome of clearly defined events that may occur at random. With fuzzy logic, we're trying to determine something about the nature of the event itself. Fuzziness is often expressed as ambiguity, not imprecision or uncertainty; it's a characteristic of perception as well as concept.

3. Designing fuzzy sets is easy. Fuzzy sets reflect, in a general way, how people actually think about a problem. It's usually quick and easy to rough out the approximate shape of a fuzzy set. Later on, after some testing or experience, we can adjust its precise characteristics.

4. Fuzzy systems are stable and easily tuned and can be validated. It's faster and easier to create fuzzy sets and build a fuzzy system than it is to create conventional knowledge-based systems, since fuzzy logic handles all the interlocking degrees of freedom. These systems are validated much like conventional systems, but tuning them is usually much simpler.

5. Fuzzy systems aren't neural networks. A fuzzy system attempts to find the intersection, union or complement of the fuzzy control variables. While this is somewhat analogous to both neural networks and linear programming, fuzzy systems approach these problems differently.

6. Fuzzy logic is more than process control. Although some people view fuzzy logic mainly as a tool for process control and signal analysis, that interpretation is too limiting. Fuzzy logic is a way of representing and analyzing information, independent of specific applications.

7. Fuzzy logic is a representational and reasoning process. Fuzzy logic is a powerful and versatile tool for representing imprecise, ambiguous and vague information. It can't solve all problems, but it helps us model difficult, even intractable problems.

Adapted from "The Seven Noble Truths of Fuzzy Logic," by Earl Cox, Computer Design, April 1992

Copyright © 2004 IDG Communications, Inc.

It’s time to break the ChatGPT habit
Shop Tech Products at Amazon