Improved Partial Least Squares Regression Recommendation Algorithm

DOI

10.2991/icaiees-13.2013.26How to use a DOI?

Keywords

partial least squares (PLS), kernel algorithm, algorithms improvement, recursive exponentially weighted algorithms

Abstract

This paper aims to improve the performance of partial least squares regression, and then, improve efficiency of its implementation. In this paper we provide a novel derivation based on optimization for the partial least squares (PLS) algorithm. The derivation shows that only one of either the X- or the Y- matrix needs to be deflated during the sequential process of computing latent factors. And then, based on this derivation, an improved recursive exponentially weighted PLS regression algorithm was proposed. And the improved algorithm is obviously superior to traditional PLS regression algorithm on performance.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Chunhua Liao
AU  - Jianqiang Du
AU  - Guohua Jin
AU  - Chunlei Chen
PY  - 2013/12
DA  - 2013/12
TI  - Improved Partial Least Squares Regression Recommendation Algorithm
BT  - Proceedings of the 2013 International Conference on Advanced Information Engineering and Education Science (ICAIEES 2013)
PB  - Atlantis Press
SP  - 91
EP  - 94
SN  - 1951-6851
UR  - https://doi.org/10.2991/icaiees-13.2013.26
DO  - 10.2991/icaiees-13.2013.26
ID  - Liao2013/12
ER  -