Chen Su-Fen
Department of Computer Science and Technology, Nanchang Institute of Technology, Nanchang Jiangxi, 330099, China
ABSTRACT
Unified Multiple Linear Regression (UMLR) is a nonlinear programming model that unifies all kind of multiple linear regression models, such as Principal Components Regression, Ridge Regression, Robust Regression and constrained regression. Although, UMLR has exhibited excellent performances in some real applications, the optimization procedure is not satisfying yet. This study proposes a novel Granular Computing-Particle Swarm Optimization (Grc-PSO) algorithm by introducing granular computing into standard PSO which is used for the optimization of the UMLR model. The experimental results show that the solution got by Grc-PSO algorithm is much better to the real situation than other state-of-art algorithms.
PDF References Citation
How to cite this article
Chen Su-Fen, 2013. Dynamic Population Structure based PSO with Granular Computing for Unified Multiple Linear Regression. Information Technology Journal, 12: 8430-8434.
DOI: 10.3923/itj.2013.8430.8434
URL: https://scialert.net/abstract/?doi=itj.2013.8430.8434
DOI: 10.3923/itj.2013.8430.8434
URL: https://scialert.net/abstract/?doi=itj.2013.8430.8434