EFFICIENT CLASSIFIER TO DETECT DDoS ATTACK BASED ON INTERNET OF THINGS

An intriguing mechanism that facilitates easy connection between several devices is the internet of things (IoT). This encourages the creation of fresh methods for automatically detecting clientIoT occurrence traffic. Through this study, we show that several kinds of machine learning methods may produce great accurateness distributed denial of service (DDoS) detection in IoT network traffic by exploiting IoT-particular network characteristics to guide choice of features. The results of the study demonstrated that our system detected DDoS attacks with high precision, confirming its dependability and robustness in IoT network. A DDoS detection algorithm that utilizes machine learning approaches is proposed in the present study. The most recent dataset, CICDDoS2019, was utilized to write this research. It tested a variety of well-liked machine learning techniques and identified the attributes that most closely correspond with projected classes. It is found that random forest was 99.5% accurate in predicting the type of network procedure, demonstrating their extraordinary accuracy.