TY - GEN
T1 - Global Optimization
T2 - 1st International Conference on Informatics and Intelligent Applications, ICIIA 2021
AU - Agbehadji, Israel Edem
AU - Awuzie, Bankole Osita
AU - Ngowi, Alfred Beati
AU - Millham, Richard C.
AU - Frimpong, Samuel Ofori
N1 - Publisher Copyright:
© 2022, Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - The social spider-prey (SSP) search is one of the newly developed search strategies for an optimization problem. SSP mimics the behaviour of the spider and prey on the spider’s web. This paper aims to propose a hybrid algorithm that avoids parameter estimation based on trial-and-error for the global optimization problem. The methodology is based on the Kestrel-based search method (KSA) to generate the best weight value of the prey in any hyper-dimensional space, which is constrained by parameters due to the nature and complexity of the problem. The proposed hybrid SSP-KSA was tested on benchmark functions through a computational experiment and the results are discussed. The results indicate that the Hybrid SSP-KSA demonstrate global optimization performance for the different dimensional waves in the search spaces. The results also show the different values of amplitude, and none reached −16 × 10−8 value. Also, the optimal value for the superimposed wave was between 0.1 and 0.13. In conclusion, irrespective of an increase in dimension space, the graph of amplitude converges to optimality which suggests a prey has finally been caught on the spider’s web.
AB - The social spider-prey (SSP) search is one of the newly developed search strategies for an optimization problem. SSP mimics the behaviour of the spider and prey on the spider’s web. This paper aims to propose a hybrid algorithm that avoids parameter estimation based on trial-and-error for the global optimization problem. The methodology is based on the Kestrel-based search method (KSA) to generate the best weight value of the prey in any hyper-dimensional space, which is constrained by parameters due to the nature and complexity of the problem. The proposed hybrid SSP-KSA was tested on benchmark functions through a computational experiment and the results are discussed. The results indicate that the Hybrid SSP-KSA demonstrate global optimization performance for the different dimensional waves in the search spaces. The results also show the different values of amplitude, and none reached −16 × 10−8 value. Also, the optimal value for the superimposed wave was between 0.1 and 0.13. In conclusion, irrespective of an increase in dimension space, the graph of amplitude converges to optimality which suggests a prey has finally been caught on the spider’s web.
KW - Hyper-dimensional search space
KW - Kestrel-based search method
KW - Social spider-prey
UR - http://www.scopus.com/inward/record.url?scp=85124661021&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-95630-1_17
DO - 10.1007/978-3-030-95630-1_17
M3 - Conference contribution
AN - SCOPUS:85124661021
SN - 9783030956295
T3 - Communications in Computer and Information Science
SP - 240
EP - 255
BT - Informatics and Intelligent Applications - 1st International Conference, ICIIA 2021, Revised Selected Papers
A2 - Misra, Sanjay
A2 - Oluranti, Jonathan
A2 - Damaševičius, Robertas
A2 - Maskeliunas, Rytis
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 25 November 2021 through 27 November 2021
ER -