Study on extension-based fuzzy pattern recognition

Pattern recognition, especially fuzzy pattern recognition, has been a study hotspot and important application of the artificial intelligence and machine learning. To solve the shortages of the existing fuzzy pattern recognition, this paper puts forward a new method, which is based on extension theory. It firstly takes advantage of the basal concepts, such as matter-element, dependent function in extension theory, to set up an extension-based fuzzy pattern recognition model, and then uses an emulational instance-the voice recognition to testify it. It is proved that this new method can not only conquer the shortages of the existing fuzzy pattern recognition methods, but also have more efficiency and applications.