Cybersecurity Challenges in Agriculture: Detecting DDoS Attacks with a Novel Machine Learning-Based Approach
T. Sethukarasi, S. Prabakeran, Babu Munirathinam, Manickam Muruganantham, Balaji Annamalai, Indumathi Varadharajan · Cybernetics & Systems · 2025
Agriculture is the greatest source for providing food to the ever-increasing human population. The various factors such as lack of skills regarding farming, augmenting demand for sustainable production of foods, and the requirement for effective and limited usage of environmental resources, put the agricultural sector into modernization. Multiple techniques have been developed for assisting farming. Still, they exhibit multiple challenges like considering security as a last priority and exposing challenges while detecting and classifying Distributed Denial of Service (DDoS) attacks. To overcome these limitations, this study presents the Ensemble Logistic Random Support vector-based Squid Game Search (ELRS-SGS) algorithm for detecting and classifying the DDoS attacks. The input data constituting multiple DDoS attack instances are gathered from the CIC-DDoS2019 dataset. The data are preprocessed, and features are extracted using the Weighted Pearson Correlation Coefficient method. Further, the attacks are recognized and classified using ensemble multiple ML methodologies such as Logistic Regression, Support Vector Machine, Random Forest, and Ensemble Voting. Moreover, the hyperparameters of these classifiers are optimized using the Squid Game Search algorithm. The results indicate a successful anomaly detection of the proposed method achieved an accuracy of 99.87% on the CIC-DDoS2019 dataset.