Slice transform-based weight updating strategy for PLS (pages 565–575)
Yiming Bi, Qiong Xie, Silong Peng and Weiying Lu
Article first published online: 30 SEP 2012 | DOI: 10.1002/cem.2479
A modified partial least squares (PLS) algorithm is presented on the basis of a novel weight updating strategy. The new weight can handle situations with directions in X space having large variance unrelated to Y, whereas the linear PLS may not work well. In the proposed algorithm, the slice transform technique is introduced to provide a piecewise linear representation of the weight vectors. An optimal piecewise linear replacements of the PLS weights are achieved by the proposed method. Experimental results show that the proposed method can achieve simpler models, whereas the model performances are at least comparable with PLS and other methods.