45 Principal Component Analysis Considering Weights Based on Dissimilarity of Objects in High Dimensional Space
Download citation file:
A weighted principal component analysis based on the result of fuzzy clustering has been proposed. The merit of this analysis is to obtain the result of the principal component analysis considered the dissimilarity structure of objects in the high dimensional space. The dissimilarity structure is represented through weights of objects which show how each object contributes to the classification structure based on the dissimilarity of objects in the high dimensional space. Since the degree of contribution is defined by a function, according to the change of the function, the weights will be changed. This paper proposes weighted principal component analysis...