EUVIMEDEUVIMED entdecken
Crossref

Rhino Optimiser Algorithm: A Novel Bio-Inspired Algorithm

Simisani Ndaba, Rajalakshmi Selvaraj, Hlomani Hlomani

Journal of Computer Science · 01.11.2026 · 22 (11) · 3387-3406

DOI: 10.3844/jcssp.2026.3387.3406 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computer Science, erschienen am 01.11.2026. Verfasst von Simisani Ndaba und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The Rhino Optimiser is a novel bio-inspired optimisation algorithm inspired by the adaptive movement behaviour of solitary rhinoceroses. Unlike conventional swarm-based metaheuristics, the proposed algorithm employs a single-agent adaptive behavioural framework designed to improve optimisation efficiency and reduce search stagnation and premature convergence in high-dimensional search spaces. The method integrates Lévy-flight exploration, momentum-guided elite exploitation, wallowing-based escape perturbation, and adaptive migration mechanisms to achieve a dynamic balance between exploration and exploitation. Mathematical modelling, computational complexity analysis, and pseudocode are provided to support theoretical transparency and reproducibility. The performance of the Rhino Optimiser was evaluated using 60 benchmark functions from the CEC 2017, CEC 2020, CEC 2021, and CEC 2022 benchmark suites under fixed evaluation conditions and multiple independent runs. Experimental results demonstrated competitive performance compared with several classical and recent bio-inspired optimisation algorithms, including Genetic Algorithm, Differential Evolution, Particle Swarm Optimisation, Grey Wolf Optimiser, Harris Hawks Optimisation, Whale Optimisation Algorithm, Marine Predators Algorithm, and Salp Swarm Algorithm. Statistical analysis confirmed significant performance differences between the compared algorithms, while convergence analysis indicated stable optimisation behaviour across diverse benchmark functions. An ablation study and a real-world transformer-based depression detection hyperparameter optimisation task further demonstrated the robustness and practical applicability of the proposed method. The findings suggest that the Rhino Optimiser provides a competitive alternative for complex continuous optimisation problems.

Zitieren

APA

Ndaba, S., Selvaraj, R., & Hlomani, H. (2026). Rhino Optimiser Algorithm: A Novel Bio-Inspired Algorithm. Journal of Computer Science, 22(11), 3387-3406. https://doi.org/10.3844/jcssp.2026.3387.3406

Vancouver

Ndaba S, Selvaraj R, Hlomani H. Rhino Optimiser Algorithm: A Novel Bio-Inspired Algorithm. Journal of Computer Science. 2026;22(11):3387-3406. doi:10.3844/jcssp.2026.3387.3406

BibTeX
@article{Ndaba2026,
  author = {Ndaba, Simisani and Selvaraj, Rajalakshmi and Hlomani, Hlomani},
  title = {Rhino Optimiser Algorithm: A Novel Bio-Inspired Algorithm},
  journal = {Journal of Computer Science},
  year = {2026},
  volume = {22},
  number = {11},
  pages = {3387--3406},
  doi = {10.3844/jcssp.2026.3387.3406}
}
Kostenlos starten →