literature pattern

Neurofuzzy Pattern Matching
(Click here for DataX™ Catalog)

About Patterns and Pattern Matching
Patterns reflect the behavioral characteristics of how a person or a system acts under certain environment. A spending pattern may represent the way a consumer spends money on different goods, such as travels, cars, or food. A defect pattern in a semiconductor equipment may indicate the way in which a part or assembly fails. By matching various patterns, a marketing specialist at a credit card company is able to better understand consumers spending habits, and therefore he can tailor his or her marketing strategies targeted to different consumer groups. By the same token, scientists study and match various patterns of machine fault in order to be able to predict and control the performance of an equipment, including advanced warning.

The Neurofuzzy Approach
The neuro-fuzzy approach is to use neural networks and fuzzy set theory to model practical systems. Zaptron has developed proprietary neuro-fuzzy technologies for pattern recognition and matching. A pattern match or recognition system is a black box constructed using multiple layers of neurons called neural networks. Neurons have the ability of memory and self-learning by training. Fuzzy logic algorithms are implemented as an inference engine which can automatically infer from facts (data).

A Case Study
One test case is given below to demonstrate the neurofuzzy pattern matching method developed by Zaptron scientists.

Questions and comments