Akademik Veri Yönetim Sistemi
International Conference on Mathematics andMathematics Education (ICMME-2025), İstanbul, Türkiye, 11 - 13 Eylül 2025, ss.119, (Özet Bildiri)
The increasing frequency and sophistication of Distributed Denial-of-Service(DDoS) attacks pose critical threats to network availability and security. This studyaims to develop an effective classification model for DDoS detection by leveragingsupervised machine learning algorithms. The CIC-DDoS2019 dataset was employed,consisting of over 400,000 traffic instances and 78 features, which were reduced to32 through preprocessing operations, including correlation analysis and featureelimination. Six widely used algorithms—Random Forest, Support Vector Machine(SVM), k-Nearest Neighbors (KNN), XGBoost, LightGBM, and CatBoost—wereimplemented to classify normal and attack traffic.Model performance was evaluatedusing Accuracy, Recall, Precision, and F1 Score, supported by confusion matrixanalyses. Experimental results revealed that CatBoost achieved the highestperformance, with an Accuracy of 98.99%, Recall of 0.8233, Precision of 0.8505, andF1 Score of 0.8341. XGBoost (Accuracy: 98.85%, F1: 0.8184) and KNN (Accuracy:98.85%, F1: 0.7957) also exhibited strong results, while SVM showed comparativelylower performance (Accuracy: 97.85%, F1: 0.6853). These findings highlight theimportance of algorithm selection, as high accuracy values may mask low recall ratesthat can undermine real-world security effectiveness. Overall, the results demonstratethat supervised machine learning algorithms can effectively detect DDoS attacks,with CatBoost particularly excelling in handling class imbalance and categoricalfeatures. The developed models can be integrated into intrusion detection systems toenhance early threat detection and mitigation. This research contributes to the fieldsof network security, data analytics, artificial intelligence, and cyber defensestrategies.