Anomaly Intrusion Detection System Based on RNA Encoding and YAKE Algorithm Using UNSW-NB15 Dataset

Billions of people use the internet daily, and the huge amount of data transmission have a significant challenge to security. Different security methods have been proposed to reduce security losses. An intrusion detection system is one of these methods, and it is a new generation of security technology that is used to monitor the system and avoid malicious activities. However, the main challenge of these systems is the variety of attackers and the repeated changing of data, which requires a flexible and active solution. To address this problem, this paper proposed a new anomaly intrusion detection system where the proposed system consists of two phases: the key extraction phase and the classification phase. In the first phase, a new RNA encoding method is proposed to encode network traffic into RNA characters, and the encoding is done by dividing the network traffic values into two groups, nominal and numerical. After that, the Yet Another Keyword Extractor method is used to extract the anomaly keys, and then a brute force algorithm is applied to classify testing dataset records as either attack or normal, where UNSW-NB15 Dataset is used as a source for the testing phase. The performance evaluation of the proposed method is done based on the detection rate, false alarm rate, and accuracy, where the obtained detection rate, false alarm rate, and accuracy are equal to 86.63 %, 4.3 %, and 88.96%. The results show that the proposed method can detect many attacks and also reduce the security risk with high efficiency.