Anomaly-based intrusion detection system based on feature selection and majority voting

Recently, cyberattacks have been more complex than in the past, as a new cyber-attack is initiated almost every day. Therefore, researchers should develop efficient intrusion detection systems (IDS) to detect cyber-attacks. In order improve and prevention of aforementioned cyber-attacks, several articles developed IDSs exploiting machine learning deep learning. this paper, way find network intrusions using combination feature selection adoptive voting investigated. NSL-KDD dataset, high-dimensional dataset that has widely used for detection, applied approach. Feature plays an important role improving accuracy testing time it eliminates less significant attributes from data set, thus saving computational power effort. The experimental results show proposed approach achieves 86.5% on test algorithm trained with selected features. addition, process each record 97.5 microseconds, which reflects model's superior performance. Comparing model existing models literature shows adaptive significantly improves accuracy, enhances efficiency, reduces false positives.