Detecting web traffic anomalies based on artificial intelligence algorithms

Nowadays, due to the increase in Internet productivity and the increasing need for Internet websites, the threats to the security of users' information on various websites have always become more serious, widespread, and dangerous. Considering the imperfect performance of intrusion detection systems in detecting attacks, improving their performance is the subject of this article. In this article, a method for anomaly detection based on machine learning and deep learning is proposed to detect attacks. This method detects anomalous and malicious traffic based on calculating the remoteness score and comparing it with the calculated threshold. The advantage of the proposed method over the presented methods is the use of algorithms for identifying remote points similar to the direction of detecting malicious traffic. These algorithms have been evaluated with the proposed method by conducting various experiments and based on the confusion matrix. The results show that the proposed method detects anomalous traffic with a detection rate of ۰.۷۱ and an accuracy of ۰.۹۷۷, and has better performance in speed and accuracy than other conventional and traditional detection methods.

Keywords:

Machine learning , deep learning , anomalous traffic , outlier detection