IoI security with elastic stack and machine learning: Intelligent defence system against Dos attacks

With the development of technology, the use of IoT devices is increasing. These devices are used in many areas such as industrial, health, transportation with capabilities such as communication and data collection. However, the increasing use of these devices also leads to an increase in cyber-attacks. In the past, examples such as Mirai botnet attacks have shown how critical the security of IoT devices is. Therefore, this study aims to develop an intelligent defense system to increase the security of IoT devices by using Elastic Stack, machine learning and deep learning technologies. The study focuses on the detection of attacks using Bot-Iot and Unsw-IoT-Botnet datasets. Random forest, Decision Tree (Information Gain), Decision Tree (Gini Index), Naive Bayes techniques were used in the study. Lstm, Rnn and Svm, which are deep learning methods, were used to detect attacks and attack types on datasets. In the designed model, %0.9998 success rate was achieved with machine learning methods and %0.98 accuracy rate was achieved with deep learning methods. With this model, it shows that artificial intelligence supported intrusion detection systems will be useful in the security of IoT devices.