Investigation of DDoS attacks in Network Intrusion Detection System Using ML Approach

Advancements in communication technology and the much improved and advanced automation technology revolutionizing the mechanization of the manufacturing industry. These have led to easy, efficient, and cost-effective manufacturing. This brand-new age is recognized as the IoT. The adoption of IoT makes life easier for many manufacturing consumers by introducing them to new services and applications, resulting in integrating connected devices as a normal practice. Despite the benefits of using IoT devices, various issues arise as network abnormalities, like DoS. This study used KNN, DT, NB, SVM, RF, LR and RF algorithms in WEKA tools to detect DDoS incursions using deep learning. These machine learning techniques are tested for DDoS detection utilizing CICDoS2020's latest record datasets. The best model was CICDDoS2020, and the KNN and RF algorithms had 99% success rates. However, based on computation speed, KNN is considered faster than RF, with computation times of 4.53 seconds & 84.2 seconds, respectively.