Host: Transdisciplinary Federation of Science and Technology
Simultaneous application of data partitioning and local modeling is a useful tool for knowledge discovery from large scale databases, and several models for local PCA and switching regression can be regarded as the hybrid of fuzzy clustering and multivariate analysis. In this research, a new approach to switching PLS regression, in which fuzzy clustering plays a role for data partitioning, is proposed by introducing fuzzy memberships into the iterative algorithm of PLS regression.