Detection of web attacks via PART classifier

With the vast and continuous growth in both computers and communications fields, despite its facilitation of work at all levels, there are a number of new challenges society is facing. The most important of which is the security of sensitive data. With so many hackers wanting to steal sensitive information and exploit it for their own unethical purposes, new protection techniques have to be found. In recent years, Intrusion Detection System (IDS) technology has emerged as an effective option for protecting information within the network. This technology can distinguish between normal traffic and intrusion within the network. In this study, the PART-machine learning classifier algorithm was used to detect web attack attempts based on one of the most recent datasets CICIDS2017. The classifier achieved more than 99% accuracy. RandomForest, NaiveBayes, and BayesNet algorithms are also tested for comparison purposes.