Search

The paper addresses the limitations of the Moth-Flame Optimization (MFO) algorithm, a meta-heuristic used to solve optimization problems. The MFO algorithm, which employs moths’ transverse orientation navigation technique, has been used to generate solutions for such problems. However, the performance of MFO is dependent on the flame production and spiral search components, and the search mechanism could still be improved concerning the diversity of flames and the moths’ ability to find solutions. The authors propose a revised version called GMSMFO, which uses a Novel Gaussian mutation mechanism and shrink MFO to enhance population diversity and balance exploration and exploitation capabilities. The study evaluates the performance of GMSMFO using the CEC 2017 benchmark and 20 datasets, including a high-dimensional intrusion detection system dataset. The proposed algorithm is compared to other advanced metaheuristics, and its performance is evaluated using statistical tests such as Friedman and Wilcoxon rank-sum. The study shows that GMSMFO is highly competitive and frequently superior to other algorithms. It can identify the ideal feature subset, improving classification accuracy and reducing the number of features used. The main contribution of this research paper includes the improvement of the exploration/exploitation balance and the expansion of the local search. The ranging controller and Gaussian mutation enhance navigation and diversity. The research paper compares GMSMFO with traditional and advanced metaheuristic algorithms on 29 benchmarks and its application to binary feature selection on 20 benchmarks, including intrusion detection systems. The statistical tests (Wilcoxon rank-sum and Friedman) evaluate the performance of GMSMFO compared to other algorithms. The algorithm source code is available at https://github.com/MohammedQaraad/GMSMFO-algorithm. [ABSTRACT FROM AUTHOR]

Particle swarm optimization (PSO) suffers from delayed convergence and stagnation in the local optimal solution, as do most meta-heuristic algorithms. This study proposes a time-based leadership particle swarm-based Salp (TPSOSA) to address the PSO's limitations. The TPSOSA is a novel search technique that addresses population diversity, an imbalance between exploitation and exploration, and the premature convergence of the PSO algorithm. Hybridization in TPSOSA is divided into two stages: The PSO hierarchy of leaders and followers is first represented as a time-varying dynamic structure. Because we need much exploration at the beginning and many exploitative steps at the end, this method raises the number of leaders while decreasing the number of follower particles linearly. In the time-varying form of the PSO (TPSOSA), unlike the PSO, the number of leaders and followers changes over time. PSO's robust search strategy is used to update the leaders' positions. Second, the SSA's powerful exploitation is utilized to update the followers' swarm population position. The purpose of tweaking the particle swarm optimizer algorithm is to aid the fundamental method in avoiding premature convergence and quickly directing the search to the most promising likely search space. The proposed TPSOSA method is tested using the CEC 2017 benchmark, seven CEC2008lsgo test functions with 200, 500, and 1000 decision variables, and 19 datasets (including three high-dimensional datasets and the NSL-KDD Dataset for Intrusion Detection System). In each experiment, TPSOSA is compared to various state-of-the-art metaheuristics methods. Friedman and Wilcoxon rank-sum statistical tests are also used to analyze the data. Experimental data and statistical tests show that the TPSOSA algorithm is very competitive and often superior to the algorithms used in the studies. According to the results, TPSOSA can also find an optimal feature subset that enhances classification accuracy while reducing the number of features employed. [ABSTRACT FROM AUTHOR]

Like most metaheuristic algorithms, salp swarm algorithm (SSA) suffers from slow convergence and stagnation in the local optima. The study develops a novel Time-Based Leadership Salp-Based Competitive Learning (TBLSBCL) to address the SSA's flaws. The TBLSBCL presents a novel search technique to address population diversity, an imbalance between exploitation and exploration, and the SSA algorithm's premature convergence. Hybridization consists of two stages: First, a time-varying dynamic structure represents the SSA hierarchy of leaders and followers. This approach increases the number of leaders while decreasing the number of salp's followers linearly. Utilizing the effective exploitation of the SSA, the position of the population's leader is updated. Second, the competitive learning strategy is used to update the status of the followers by teaching them from the leaders. The goal of adjusting the salp swarm optimizer algorithm is to help the basic approach avoid premature convergence and quickly steer the search to the most promising likely search space. The proposed TBLSBCL method is tested using the CEC 2017 benchmark, feature selection problems for 19 datasets (including three high-dimensional datasets). The TBLSBCL was then evaluated using a benchmark set of seven wellknown constrained design challenges in diverse engineering fields defined in the benchmark set of real-world problems presented at the CEC 2020 conference (CEC 2020). In each experiment, TBLSBCL is compared with seven other state-of-the-art metaheuristics and other advanced algorithms that include seven variants of the salp swarm. Friedman and Wilcoxon rank-sum statistical tests are also used to examine the results. According to the experimental data and statistical tests, the TBLSBCL algorithm is very competitive and often superior to the algorithms employed in the studies. [ABSTRACT FROM AUTHOR]