Optimization of Support Vector Regression Parameters by Flower Pollination Algorithm

DOI

10.2991/fmsmt-17.2017.313How to use a DOI?

Keywords

Support Vector Regression; Parameters Optimization; Flower Pollination Algorithm; GA; PSO.

Abstract

Support vector regression (SVR) is widely applied as a powerful method for data regression in engineering design and optimization. The regression accuracy and generalization performance of SVR model depend on the proper setting of its parameters. To this end, it is necessary to find an automated reliable, accurate and robust optimization approach to determining the optimal SVR parameter setting. This paper presents a SVR parameters optimization approach based on flower pollination algorithm (FPA), termed as FPA-SVR, to enhance the prediction ability of SVR model. Then a comparison is made among the performance of GA-SVR, PSO-SVR and FPA-SVR on one standard dataset. It can be concluded from the numerical results that the FPA-SVR model has superior regression accuracy and generalization performance.

Copyright

© 2017, 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  - Yuan Yang
AU  - Zhongqi Wang
AU  - Bo Yang
AU  - Xudong Liu
PY  - 2017/04
DA  - 2017/04
TI  - Optimization of Support Vector Regression Parameters by Flower Pollination Algorithm
BT  - Proceedings of the 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017)
PB  - Atlantis Press
SP  - 1607
EP  - 1612
SN  - 2352-5401
UR  - https://doi.org/10.2991/fmsmt-17.2017.313
DO  - 10.2991/fmsmt-17.2017.313
ID  - Yang2017/04
ER  -