Sign in Subscribe
Machine learning

Machine Learning Breakthrough: Improved Tyrannosaurus Optimization Algorithm Achieves High Accuracy in Abnormal Traffic Detection

Strategic Insights

— 1 min read

A team of researchers at Hubei University of Technology, led by Jiahui Chen, has made significant strides in machine learning by developing an improved version of the Tyrannosaurus Optimization Algorithm (TROA). The new algorithm, known as ITROA, has been successfully applied to abnormal traffic detection for Software-Defined Networking (SDN), achieving an impressive accuracy rate of 99.37% on binary classification and 96.73% on multiclassification. The research, funded by the National Natural Science Foundation of China (NSFC) and the Hubei Provincial Science and Technology Plan Project, has been peer-reviewed and published in the Cmc-computers Materials & Continua journal.

Key Takeaways:

  • The researchers introduced the Metaheuristic Algorithm (MA) to select features before machine learning, reducing the dimensionality of data and improving the efficiency of abnormal traffic detection.
  • The Improved Tyrannosaurus Optimization Algorithm (ITROA) was proposed as a metaheuristic-driven abnormal traffic detection model for SDN, which outperformed other machine learning algorithms in terms of accuracy and convergence speed.
  • The experiment showed that ITROA achieved an accuracy of 99.37% on binary classification and 96.73% on multiclassification, outperforming other compared machine learning algorithms.
  • The researchers applied ITROA to a benchmark function and the UCI dataset to verify its validity, and performed feature selection optimization operation on the InSDN dataset to obtain the optimized feature subset for SDN abnormal traffic detection.
  • The study highlighted the advantages of ITROA, including few parameters, simple implementation, and fast convergence, making it a promising solution for abnormal traffic detection in SDN.

Statistics:

  • The accuracy rate of ITROA achieved on binary classification was 99.37%.
  • The accuracy rate of ITROA achieved on multiclassification was 96.73%.
  • The experiment showed that ITROA outperformed other compared machine learning algorithms in terms of accuracy and convergence speed.
  • The study involved the National Natural Science Foundation of China (NSFC) and the Hubei Provincial Science and Technology Plan Project as funders.
  • The research was conducted at Hubei University of Technology, School of Computer Science, Wuhan, People's Republic of China.

Sources:

  • Metaheuristic-driven Abnormal Traffic Detection Model for Sdn Based On Improved Tyrannosaurus Optimization Algorithm. Cmc-computers Materials & Continua, 2025;83(3):4495-4513.
  • Hubei University of Technology, School of Computer Science, Wuhan 430068, People's Republic of China.
  • National Natural Science Foundation of China (NSFC).
  • Hubei Provincial Science and Technology Plan Project.
  • Cmc-computers Materials & Continua can be contacted at: Tech Science Press, 871 Coronado Center Dr, Suite 200, Henderson, NV 89052, USA.

Read more