期刊名称:International Journal of Computer and Information Technology
印刷版ISSN:2279-0764
出版年度:2018
卷号:7
期号:2
页码:55
出版社:International Journal of Computer and Information Technology
摘要:Genetic algorithm (GA) is a rapidly growing area ofArtificial Intelligence. This paper introduces a faster and lesscomputationally expensive enhanced Genetic Algorithm (GA)than the standard GA. Five enhancements are introduced here(multiple weighted roulettes, multiple cross over points, multiplemates, utilizing the D4 wavelets and using normal distributionfor selecting the initial population). Then the new enhanced GAwas applied to a dynamic large optimization problem that usessymbolic data to differentiate between edible and poisonousmushrooms using twenty two different characteristics. Resultswas obtained using the enhanced GA as well as the standard GA.The enhanced GA was able to obtain the results using lessnumber of calculations, which indicates it was lesscomputationally expensive. A special java program wasdeveloped for this purpose; and Excel was used to represent thecharts data.
关键词:genetic algorithm; bio inspired system; evolutionary;algorithm; weighted roulette wheel; Daubechies wavelets;mushrooms; symbolic data; normal distribution