Classification of Distributed Denial Service (DDoS) Attacks Using the K-Nearest Neighbor (KNN) Method

Distributed Denial of Service (DDoS) attacks pose a significant security threat by disrupting network services through a flood of data traffic or exploiting system vulnerabilities. Early detection of DDoS attacks is essential to reduce their potential impact. This study aims to classify DDoS attacks using the K-Nearest Neighbor (KNN) algorithm to improve network security. The research data is sourced from a publicly available DDoS Software-Defined Networking (SDN) dataset. The research stages include data collection, pre-processing, implementation of the KNN algorithm, and model evaluation. Data pre-processing involves data cleansing, feature transformation, and normalization to optimize model performance. The KNN algorithm determines the number of Ks of the nearest neighbor and uses geometric distances to classify DDoS attacks. The conclusion of this study assesses the accuracy of the KNN model in detecting DDoS attacks. The results of the evaluation showed that the KNN model reached a level of accuracy