Support vector machine kernel function selection method under sparse representation and application thereof

A support vector machine and sparse representation technology, applied in complex mathematical operations, computer components, instruments, etc., to achieve the effect of strong generalization ability

Inactive Publication Date: 2015-03-25
JIANGXI UNIV OF SCI & TECH
View PDF6 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The traditional SVM kernel function selection method is artifi

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
  • Support vector machine kernel function selection method under sparse representation and application thereof
  • Support vector machine kernel function selection method under sparse representation and application thereof
  • Support vector machine kernel function selection method under sparse representation and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Embodiment 1: Sewage detection experimental data

[0038] Step A: 57 sets of sample data were collected in a sewage treatment plant and preprocessed to make the norm of the sample data less than 1; then DO (dissolved oxygen), BOD5, membrane pressure difference, and water production in the sewage were selected The seven parameters of daily treatment capacity, actual residence time, and influent load are used as input variables, and COD and NH4+-N indicators in the effluent treated by membrane biotechnology are used as output variables;

[0039] Step B: select and construct different kernel functions that meet the Mercer conditions, and establish a complete sparse dictionary of SVM kernel functions;

[0040] Step C: select 42 sets of sample data as training samples, 15 sets of sample data as test samples, and then use the sparse representation theory to obtain the sparse coding of the sample data under the sparse dictionary;

[0041] Step D: select the SVM kernel functio...

Embodiment 2

[0047] Embodiment 2: Servo Data Set (servo data set)

[0048] Step A: This data set comes from the UCI database, with a total of 167 sets of data, and it is preprocessed to make the norm of the sample data less than 1; then select 4 motor types, screw specifications, amplification factor p, and amplification factor v is the input variable, and the servo motor type is the output variable;

[0049] Step B: select and construct different kernel functions that meet the Mercer conditions, and establish a complete sparse dictionary of SVM kernel functions;

[0050] Step C: select 132 sets of sample data as training samples, 35 sets of sample data as test samples, and then use the sparse representation theory to obtain the sparse coding of the sample data under the sparse dictionary;

[0051] Step D: select SVM kernel function type number n=1 through sparse coding, SVM kernel function type RBF radial basis kernel function k RBF , at this time ρ=1;

[0052] E step: utilize particle s...

Embodiment 3

[0054] Embodiment three: Yacht Hydrodynamics Data Set (yacht hydrodynamics data set)

[0055] Step A: This data set comes from the UCI database, with a total of 308 sets of data, which are used to predict the hydrodynamic performance of the yacht, the size and speed of the sailing yacht; preprocess it so that the norm of the sample data is less than 1; then select the center of buoyancy Six input variables are longitudinal position, rhombus coefficient, ship length displacement ratio, ship width displacement ratio, ship length width ratio, and Froude coefficient, and the displacement per unit weight of residual resistance is output variable;

[0056] Step B: select and construct different kernel functions that meet the Mercer conditions, and establish a complete sparse dictionary of SVM kernel functions;

[0057] Step C: Select 240 sets of sample data as training samples, 68 sets of sample data as test samples, and then use the sparse representation theory to obtain the sparse...

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 provides a novel support vector machine kernel function selection method in which a sparse representation theory is applied to support vector machine kernel function selection and application. The method includes the steps that (1) specific sample data are given and pre-processed, (2) an SVM kernel function sparse dictionary meeting Mercer conditions is selected and built, (3) a sparse code is solved, (4) an SVM kernel function type is selected according to the solved sparse code, (5) corresponding SVM parameters are optimized and a support vector machine model is determined, and (6) a prediction result is output. Attribution representation and modeling ability of sample data are achieved by means of the sparse theory, sample data prior information of actual problems is effectively utilized, metric characteristics of different kernel functions are taken into consideration for SVM modeling, generalization ability is high, and the defect that in a traditional SVM model selection method, the kernel function type is manually assigned and accordingly models can not have the optimal performance is overcome.

Description

technical field [0001] The invention applies the sparse representation theory to the kernel function selection of the support vector machine, and is a new method and application of the kernel function selection of the support vector machine. Background technique [0002] Support Vector Machine (SVM) is a new machine learning method proposed by Vapnik based on statistical learning theory in the 1990s. SVM is a learning method based on kernel, the selection of kernel function has an important influence on the performance of support vector machine, how to effectively select kernel function is an important issue in the field of SVM research. Due to the different geometric metric features contained in different kernel functions, the selection of different kernel functions leads to differences in the generalization ability of SVM. Therefore, for specific practical problems, it is very important to choose what kind of kernel function, and it is also a major problem encountered in ...

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): G06F17/15G06K9/62
Inventor 梁礼明钟震杨国亮冯新刚林元璋袁晓
Owner JIANGXI UNIV OF SCI & TECH
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