SVM classifier parameter optimization method based on improved particle swarm algorithm

A technology for improving particle swarm optimization and optimization methods, applied in instruments, calculations, calculation models, etc., can solve problems such as low comprehensive performance of classifiers and poor generalization ability of classifiers, and achieve enhanced generalization ability, enhanced flexibility, and high The effect of classification accuracy

Inactive Publication Date: 2018-11-23
SOUTHEAST UNIV
View PDF0 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides a method for optimizing parameters of an SVM classifier based

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • SVM classifier parameter optimization method based on improved particle swarm algorithm

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0028] Such as figure 1 As described, this method first selects the parameters to be optimized, first collects sample data, and performs 10-fold cross-validation experiments on the sample data. During the experiment, one parameter of the classifier is selected as an independent variable, and other parameters of the classifier are fixed. After the experiment is completed, compare the impact of each parameter on the performance of the classifier from the three indicators of average classification accuracy, test accuracy and support vector ratio, and select several parameters that have a greater impact on the performance of the classifier as the parameters to be optimized.

[0029] After experiments, four parameters to be optimized are selected, namely the polynomial kernel function weight a, the polynomial kernel parameter d, and the Gaussian kernel parameter g (g=σ 2 ) And the penalty factor C.

[0030] Then to the parameter optimization stage, including:

[0031] (1) Initialize th...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a SVM classifier parameter optimization method based on an improved particle swarm algorithm. The method comprises the following steps: (1) performing 10-fold cross verification on collected sample data, selecting the parameter influencing the classifier performance as the to-be-optimized parameter; (2) initializing related parameters of the classifier and the particle swarm algorithm, and updating particle speed and location according to related parameters; (3) setting the to-be-optimized parameter of the classifier as the corresponding dimension value at the current location of the particle, and computing to obtain the fitness value corresponding to the current location of the particle; and (4) obtaining the fitness value evaluation particle according to the fitness value corresponding to the current location of the particle, and updating the individual optimal location and the population optimal location. Through the method disclosed by the invention, the mixed kernel function based on the polynomial kernel function and the Gaussian kernel function is constructed, the traditional particle swarm algorithm is improved, the kernel function parameter is optimized by utilizing the improved PSO-SVM algorithm, and then the comprehensive performance of the classifier is improved, the generalization capacity of the classifier is improved when the high classification precision is guaranteed.

Description

technical field [0001] The invention relates to a parameter optimization method of an SVM classifier, in particular to a parameter optimization method of an SVM classifier based on an improved particle swarm algorithm. Background technique [0002] Kernel function is the key factor affecting the performance of SVM classifier. Usually we divide kernel functions into two categories: local kernel functions and global kernel functions. The former has weak generalization ability and strong learning ability, while the latter has strong generalization ability and weak learning ability. Among the commonly used kernel functions, the polynomial kernel function belongs to the global kernel function, and the Gaussian kernel function belongs to the local kernel function. The selection of the kernel function includes two steps: the determination of the kernel function and the determination of the parameter value of the kernel function. [0003] For the determination of the kernel funct...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/62G06N3/00
CPCG06N3/006G06F18/2411
Inventor 黄杰周微
Owner SOUTHEAST UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap