Feature Selection and Parameter Optimization of Support Vector Machines Based on a Local Search Based Firefly Algorithm for Classification of Formulas in Traditional Chinese Medicine

Wen SHI; Jianling LIU; Jingyu ZHANG; Yuran MEN; Hongwei CHEN; Deke WANG; Yang CAO

doi:10.1587/transfun.2021EAL2075

Open Access
Feature Selection and Parameter Optimization of Support Vector Machines Based on a Local Search Based Firefly Algorithm for Classification of Formulas in Traditional Chinese Medicine

Wen SHI, Jianling LIU, Jingyu ZHANG, Yuran MEN, Hongwei CHEN, Deke WANG, Yang CAO

Full Text Views

118

Cite this

Free PDF (565.3KB)

Summary :

Syndrome is a crucial principle of Traditional Chinese Medicine. Formula classification is an effective approach to discover herb combinations for the clinical treatment of syndromes. In this study, a local search based firefly algorithm (LSFA) for parameter optimization and feature selection of support vector machines (SVMs) for formula classification is proposed. Parameters C and γ of SVMs are optimized by LSFA. Meanwhile, the effectiveness of herbs in formula classification is adopted as a feature. LSFA searches for well-performing subsets of features to maximize classification accuracy. In LSFA, a local search of fireflies is developed to improve FA. Simulations demonstrate that the proposed LSFA-SVM algorithm outperforms other classification algorithms on different datasets. Parameters C and γ and the features are optimized by LSFA to obtain better classification performance. The performance of FA is enhanced by the proposed local search mechanism.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E105-A No.5 pp.882-886

Publication Date: 2022/05/01

Publicized: 2021/11/16

Online ISSN: 1745-1337

DOI: 10.1587/transfun.2021EAL2075

Type of Manuscript: LETTER

Category: Algorithms and Data Structures

Authors

Wen SHI
  the Tianjin University of Commerce
Jianling LIU
  the Tianjin University of Commerce
Jingyu ZHANG
  the Tianjin Nankai Hospital
Yuran MEN
  the Tianjin University of Commerce
Hongwei CHEN
  the Tianjin University of Commerce
Deke WANG
  the Tianjin University of Commerce
Yang CAO
  the Tianjin University of Commerce

Keyword

classification, firefly algorithm, feature selection, parameter optimization

Cite this

Copy

Wen SHI, Jianling LIU, Jingyu ZHANG, Yuran MEN, Hongwei CHEN, Deke WANG, Yang CAO, "Feature Selection and Parameter Optimization of Support Vector Machines Based on a Local Search Based Firefly Algorithm for Classification of Formulas in Traditional Chinese Medicine" in IEICE TRANSACTIONS on Fundamentals, vol. E105-A, no. 5, pp. 882-886, May 2022, doi: 10.1587/transfun.2021EAL2075.
Abstract: Syndrome is a crucial principle of Traditional Chinese Medicine. Formula classification is an effective approach to discover herb combinations for the clinical treatment of syndromes. In this study, a local search based firefly algorithm (LSFA) for parameter optimization and feature selection of support vector machines (SVMs) for formula classification is proposed. Parameters C and γ of SVMs are optimized by LSFA. Meanwhile, the effectiveness of herbs in formula classification is adopted as a feature. LSFA searches for well-performing subsets of features to maximize classification accuracy. In LSFA, a local search of fireflies is developed to improve FA. Simulations demonstrate that the proposed LSFA-SVM algorithm outperforms other classification algorithms on different datasets. Parameters C and γ and the features are optimized by LSFA to obtain better classification performance. The performance of FA is enhanced by the proposed local search mechanism.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2021EAL2075/_p

Copy

@ARTICLE{e105-a_5_882,
author={Wen SHI, Jianling LIU, Jingyu ZHANG, Yuran MEN, Hongwei CHEN, Deke WANG, Yang CAO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Feature Selection and Parameter Optimization of Support Vector Machines Based on a Local Search Based Firefly Algorithm for Classification of Formulas in Traditional Chinese Medicine},
year={2022},
volume={E105-A},
number={5},
pages={882-886},
abstract={Syndrome is a crucial principle of Traditional Chinese Medicine. Formula classification is an effective approach to discover herb combinations for the clinical treatment of syndromes. In this study, a local search based firefly algorithm (LSFA) for parameter optimization and feature selection of support vector machines (SVMs) for formula classification is proposed. Parameters C and γ of SVMs are optimized by LSFA. Meanwhile, the effectiveness of herbs in formula classification is adopted as a feature. LSFA searches for well-performing subsets of features to maximize classification accuracy. In LSFA, a local search of fireflies is developed to improve FA. Simulations demonstrate that the proposed LSFA-SVM algorithm outperforms other classification algorithms on different datasets. Parameters C and γ and the features are optimized by LSFA to obtain better classification performance. The performance of FA is enhanced by the proposed local search mechanism.},
keywords={},
doi={10.1587/transfun.2021EAL2075},
ISSN={1745-1337},
month={May},}

Copy

TY - JOUR
TI - Feature Selection and Parameter Optimization of Support Vector Machines Based on a Local Search Based Firefly Algorithm for Classification of Formulas in Traditional Chinese Medicine
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 882
EP - 886
AU - Wen SHI
AU - Jianling LIU
AU - Jingyu ZHANG
AU - Yuran MEN
AU - Hongwei CHEN
AU - Deke WANG
AU - Yang CAO
PY - 2022
DO - 10.1587/transfun.2021EAL2075
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E105-A
IS - 5
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - May 2022
AB - Syndrome is a crucial principle of Traditional Chinese Medicine. Formula classification is an effective approach to discover herb combinations for the clinical treatment of syndromes. In this study, a local search based firefly algorithm (LSFA) for parameter optimization and feature selection of support vector machines (SVMs) for formula classification is proposed. Parameters C and γ of SVMs are optimized by LSFA. Meanwhile, the effectiveness of herbs in formula classification is adopted as a feature. LSFA searches for well-performing subsets of features to maximize classification accuracy. In LSFA, a local search of fireflies is developed to improve FA. Simulations demonstrate that the proposed LSFA-SVM algorithm outperforms other classification algorithms on different datasets. Parameters C and γ and the features are optimized by LSFA to obtain better classification performance. The performance of FA is enhanced by the proposed local search mechanism.
ER -