2009Jisuanji gongcheng yu shejiRequires access

Comparisons of selection strategy in genetic algorithm

Zhi‐Hui Zhan

Open publisher page 13 citations

Abstract

The roulette wheel selection strategy and tournament selection strategy in genetic algorithm(GA) are taken as examples and their performance is investigated on 13 benchmark functions.The performance of different selection strategies are compared and analyzed.Experimental results show that tournament selection strategy is more general than roulette wheel selection strategy, and also with better performance.Further experiments on tournament selection strategy show that a group scale with 60% to 80% of the population size performs better.These results give the useful guideline to design more efficient selection strategy.

About this research paper

What this paper is about

The roulette wheel selection strategy and tournament selection strategy in genetic algorithm(GA) are taken as examples and their performance is investigated on 13 benchmark functions.The performance of different selection strategies are compared and analyzed.Experimental results show that tournament selection strategy is more general than roulette wheel selection strategy, and also with better performance.Further experiments on tournament selection strategy show that a group scale with 60% to 80% of the population size performs better.These results give the useful guideline to design more efficient selection strategy.

Why it matters

OpenAlex reports 13 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The roulette wheel selection strategy and tournament selection strategy in genetic algorithm(GA) are taken as examples and their performance is investigated on 13 benchmark functions.The performance of different selection strategies are compared and analyzed.Experimental results show that tournament selection strategy is more general than roulette wheel selection strategy, and also with better performance.Further experiments on tournament selection strategy show that a group scale with 60% to 80% of the population size performs better.These results give the useful guideline to design more efficient selection strategy.

Key concepts: Fitness proportionate selection, Tournament selection, Roulette, Selection (genetic algorithm), Tournament, Computer science, Benchmark (surveying), Genetic algorithm

Related papers

Back to paper searchBrowse research topicsOriginal source
Comparisons of selection strategy in genetic algorithm — Research Paper | ScholarLens