Weighted principal component analysis for interval-valued data based on fuzzy clustering
Mika Sato‐Ilic · 2004
This paper proposes a weighted principal component analysis (WPCA) for interval-valued data using the result of fuzzy clustering. In this method, we introduce two data structures which are classification structure and principal component structure. One of them is used for weights and the other is used for the analysis of itself. So, we can reduce the risk of a wrong assumption of the introduced data structure, comparing the conventional method which assumes only one data structure on the observation.