TY - JOUR
T1 - Bio-inspired computation
T2 - Recent development on the modifications of the cuckoo search algorithm
AU - Chiroma, Haruna
AU - Herawan, Tutut
AU - Fister, Iztok
AU - Fister, Iztok
AU - Abdulkareem, Sameem
AU - Shuib, Liyana
AU - Hamza, Mukhtar Fatihu
AU - Saadi, Younes
AU - Abubakar, Adamu
N1 - Publisher Copyright:
© 2017 Elsevier B.V.
PY - 2017/12
Y1 - 2017/12
N2 - Presently, the Cuckoo Search algorithm is attracting unprecedented attention from the research community and applications of the algorithm are expected to increase in number rapidly in the future. The purpose of this study is to assist potential developers in selecting the most suitable cuckoo search variant, provide proper guidance in future modifications and ease the selection of the optimal cuckoo search parameters. Several researchers have attempted to apply several modifications to the original cuckoo search algorithm in order to advance its effectiveness. This paper reviews the recent advances of these modifications made to the original cuckoo search by analyzing recent published papers tackling this subject. Additionally, the influences of various parameter settings regarding cuckoo search are taken into account in order to provide their optimal settings for specific problem classes. In order to estimate the qualities of the modifications, the percentage improvements made by the modified cuckoo search over the original cuckoo search for some selected reviews studies are computed. It is found that the population reduction and usage of biased random walk are the most frequently used modifications. This study can be used by both expert and novice researchers for outlining directions for future development, and to find the best modifications, together with the corresponding optimal setting of parameters for specific problems. The review can also serve as a benchmark for further modifications of the original cuckoo search.
AB - Presently, the Cuckoo Search algorithm is attracting unprecedented attention from the research community and applications of the algorithm are expected to increase in number rapidly in the future. The purpose of this study is to assist potential developers in selecting the most suitable cuckoo search variant, provide proper guidance in future modifications and ease the selection of the optimal cuckoo search parameters. Several researchers have attempted to apply several modifications to the original cuckoo search algorithm in order to advance its effectiveness. This paper reviews the recent advances of these modifications made to the original cuckoo search by analyzing recent published papers tackling this subject. Additionally, the influences of various parameter settings regarding cuckoo search are taken into account in order to provide their optimal settings for specific problem classes. In order to estimate the qualities of the modifications, the percentage improvements made by the modified cuckoo search over the original cuckoo search for some selected reviews studies are computed. It is found that the population reduction and usage of biased random walk are the most frequently used modifications. This study can be used by both expert and novice researchers for outlining directions for future development, and to find the best modifications, together with the corresponding optimal setting of parameters for specific problems. The review can also serve as a benchmark for further modifications of the original cuckoo search.
KW - Convergence
KW - Cuckoo search algorithm
KW - Nature-inspired algorithms
KW - Optimization metaheuristics
KW - Swarm intelligence
UR - http://www.scopus.com/inward/record.url?scp=85030870893&partnerID=8YFLogxK
U2 - 10.1016/j.asoc.2017.07.053
DO - 10.1016/j.asoc.2017.07.053
M3 - Review article
AN - SCOPUS:85030870893
SN - 1568-4946
VL - 61
SP - 149
EP - 173
JO - Applied Soft Computing
JF - Applied Soft Computing
ER -