Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Adaptive mutation particle swarm optimization algorithm

A technology of mutation particle swarm and optimization algorithm, applied in computing, computer parts, instruments, etc., can solve the problems of premature convergence and time-consuming of basic particle swarm algorithm, and achieve the effect of fast convergence, less time consumption, and improved accuracy.

Inactive Publication Date: 2017-01-25
DALIAN UNIV OF TECH
View PDF0 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to avoid the time-consuming global search of grid division and the problem that the basic particle swarm optimization algorithm is easy to fall into premature convergence (that is, local optimum), an adaptive mutation particle swarm optimization algorithm is proposed.

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
  • Adaptive mutation particle swarm optimization algorithm
  • Adaptive mutation particle swarm optimization algorithm
  • Adaptive mutation particle swarm optimization algorithm

Examples

Experimental program
Comparison scheme
Effect test

experiment example

[0035] In matlab, we first use the basic particle swarm optimization algorithm to optimize the parameters in the constructed classifier. Here, we first need to initialize the parameters of the ordinary PSO function, and set the parameter local search capability c 1 = 1.5, parameter global search ability c 2 =1.7; set the elastic coefficient ω=1 in front of the speed in the speed update formula; set the maximum number of evolution maxgen=200, the maximum number of population sizepop=20; set the number of folds of cross-validation v=5; finally set the variation range of the parameter c in Between [0.1, 100], the variation range of the parameter g is between [0.01, 1000].

[0036] First, use the basic particle swarm optimization function to find the best parameters of the SVM classifier, and the fitness curve is as follows: figure 2 Shown:

[0037] Using the adaptive mutation particle swarm algorithm to find the best parameters of the SVM classifier, the fitness curve is as f...

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 an adaptive mutation particle swarm optimization algorithm. According to the adaptive mutation particle swarm optimization algorithm, a mutation operation is introduced into PSO (Particle Swarm Optimization), namely a mutation probability factor is introduced into a whole swarm position; the mutation operation usually serves as a trigger applied to generation stagnancy (premature convergence) under control of each generation or a prefix interval or an adaptive strategy. The algorithm is capable of jumping out the current searched local optimal position and searching again in a larger solution space, so that the solution space searching range is expanded while the swarm diversity is retained; the algorithm is capable of effectively carrying out global search, so that the capability of searching global optimal solution of the swarm can be improved. The adaptive mutation particle swarm optimization algorithm is capable of finding the global optimal solution to a maximum extent under the precondition that the global search is not carried out; and the classifier performance can be improved.

Description

technical field [0001] This algorithm belongs to the field of data mining and involves SVM classifier and parameter optimization, especially its mutation operation, so that the global optimal solution can be searched to the greatest extent. Background technique [0002] The grid division method is a parameter optimization method commonly used in SVM classifiers for the following reasons: First, the grid division method is a global search, we can get the global optimal solution, and we will feel safer psychologically compared to A heuristic method that avoids exhaustive parameter search is used; the second is that finding the optimal parameter values ​​by meshing does not take more time than those advanced algorithms, since there are only two parameters. Furthermore, meshing is easy to parallelize because each pair (C, g) is independent of each other. However, sometimes in practical application problems, we need to search for the best parameters C and g on a very large inter...

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): G06K9/62
CPCG06F18/2155G06F18/2411
Inventor 吴迪高晨宵王冬伟
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com