A new algorithm of the fuzzy clustering analysis in the fuzzy pattern recognition

In the fuzzy clustering analysis, this paper puts forward a new algorithm to determine two objects similarity degree, the new algorithm is that the Angle cosine method and the hamming distance method bond together based on a weight synthesis. The characteristic of similarity algorithm is that it can adjust the weight coefficient according to the size of values calculated by the Angle cosine method and the hamming distance method alone; In the fuzzy pattern recognition, in order to avoid happening of “miss recognition”, the paper also proposes a new kind of closeness degree which is combined by two closeness degree based on a certain weight, and it can adaptively adjust each weight according to the characteristics of the characteristic value of mode; Finally, a typical application example of fuzzy clustering analysis in the fuzzy pattern recognition is presented to illustrate the effectiveness of the two new algorithms.