Feature selection algorithm based on SVM

被引：0

作者：

Sun Jiongjiong ^{[1
]}

Liu Jun ^{[1
]}

Wei Xuguang ^{[2
]}

机构：

[1] Fundamental Sci Commun Informat Transmiss & Fus T, Hangzhou 310018, Zhejiang, Peoples R China

[2] Northeastern Univ, Engn & Res Inst Co Ltd, Guiyang Off, Guiyang 550000, Peoples R China

来源：

PROCEEDINGS OF THE 35TH CHINESE CONTROL CONFERENCE 2016 | 2016年

关键词：

Feature selection; SFFS; SVM classifier; cross-validation; IMAGE RETRIEVAL; CLASSIFICATION;

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Present feature selection algorithm is very complex and not accurate enough. So this paper uses SFFS (Sequential Floating Forward Selection) algorithm to implement feature selection of target image. SFFS includes two main steps: forward selection and backtrack, it is able to avoid local optimal solution. The judgment basis of two steps is the classification accuracy of target images. SVM (Support Vector Machine) classifier obtains classification accuracy quickly and accurately by training and testing target images. To increase accuracy, this paper uses cross-validation during the classification process.

引用

页码：4113 / 4116

页数：4

共 12 条

[1] Feature extraction approaches from natural language requirements for reuse in software product lines: A systematic literature review [J].

Bakar, Noor Hasrina ;

Kasirun, Zarinah M. ;

Salleh, Norsaremah .

JOURNAL OF SYSTEMS AND SOFTWARE, 2015, 106 :132-149

[2] A review of microarray datasets and applied feature selection methods [J].

Bolon-Canedo, V. ;

Sanchez-Marono, N. ;

Alonso-Betanzos, A. ;

Benitez, J. M. ;

Herrera, F. .

INFORMATION SCIENCES, 2014, 282 :111-135

[3] A fast and effective method to find correlations among attributes in databases [J].

de Sousa, Elaine P. M. ;

Traina, Caetano, Jr. ;

Traina, Agma J. M. ;

Wu, Leejay ;

Faloutsos, Christos .

DATA MINING AND KNOWLEDGE DISCOVERY, 2007, 14 (03) :367-407

[4] Long-term streamflow forecasts by Adaptive Neuro-Fuzzy Inference System using satellite images and K-fold cross-validation (Case study: Dez, Iran) [J].

Esmaeelzadeh, Seyed Reza ;

Adib, Arash ;

Alahdin, Soroosh .

KSCE JOURNAL OF CIVIL ENGINEERING, 2015, 19 (07) :2298-2306

[5] Efficient greedy feature selection for unsupervised learning [J].

Farahat, Ahmed K. ;

Ghodsi, Ali ;

Kamel, Mohamed S. .

KNOWLEDGE AND INFORMATION SYSTEMS, 2013, 35 (02) :285-310

[6] Decision Fusion of Multiple Classifiers for Corollary Plaque Characterization from IVUS Images [J].

Giannoglou, V. G. ;

Theocharis, J. B. .

INTERNATIONAL JOURNAL ON ARTIFICIAL INTELLIGENCE TOOLS, 2014, 23 (03)

[7] Study of image retrieval and classification based on adaptive features using genetic algorithm feature selection [J].

Lin, Chuen-Horng ;

Chen, Huan-Yu ;

Wu, Yu-Shung .

EXPERT SYSTEMS WITH APPLICATIONS, 2014, 41 (15) :6611-6621

[8] Mass Classification in Mammograms Using Selected Geometry and Texture Features, and a New SVM-Based Feature Selection Method [J].

Liu, Xiaoming ;

Tang, Jinshan .

IEEE SYSTEMS JOURNAL, 2014, 8 (03) :910-920

[9] Bayesian reconstruction of multiscale local contrast images from brain activity [J].

Song, Sutao ;

Ma, Xinyue ;

Zhan, Yu ;

Zhan, Zhichao ;

Yao, Li ;

Zhang, Jiacai .

JOURNAL OF NEUROSCIENCE METHODS, 2013, 220 (01) :39-45

[10] Genetic algorithms in feature and instance selection [J].

Tsai, Chih-Fong ;

Eberle, William ;

Chu, Chi-Yuan .

KNOWLEDGE-BASED SYSTEMS, 2013, 39 :240-247

← 1 2 →