A Modified Bio Inspired: BAT Algorithm

• Published in: International journal of applied metaheuristic computing Vol. 9; no. 1; pp. 60 - 77
• Main Author: Singh, Dharmpal
• Format: Journal Article
• Language: English
• Published: Hershey IGI Global 01.01.2018
• Subjects: Accuracy; Algorithms; Analysis; Artificial intelligence; Benchmarks; Breast cancer; Chi-square test; Concrete; Data mining; Datasets; Diabetes; Echolocation (Physiology); Heuristic methods; Integer programming; Investment analysis; Knowledge discovery; Machine learning; Optimization; Particle swarm optimization; Search algorithms; Statistical methods; Swarm intelligence; Wisconsin
• ISSN: 1947-8283; 1947-8291
• DOI: 10.4018/IJAMC.2018010105

Metaheuristics algorithms are becoming powerful methods for solving many problems of market analysis, data mining, transportation, medical etc. The concept of BAT algorithm, particle swarm optimization, artificial bee colony optimization, cuckoo search, firefly algorithm and harmony search are powerful methods for solving many optimization problems. Here, an effort has been made to propose as modified form of the BAT algorithm based natural echolocation behaviour of bats to solve the optimization problems. The algorithm is also compared other 15 existing benchmark algorithms including statistical methods on five benchmarks data sets. Furthermore, modified BAT algorithm has outperformed the other algorithm in term of robustness and efficiency. The optimality of the algorithm has been also crosscheck with residual analysis and chi (χ2) square testing.