Using data analytics in the problem of Intrusion Detection and Prevention Systems (IDS/IPS) is a continuous research problem due to the evolutionary nature of the problem and the changes in major influencing factors. The main challenges in this area are designing rules that can predict malware in unknown territories and dealing with the complexity of the problem and the conflicting requirements regarding high accuracy of detection and high efficiency. In this scope, we evaluated the usage of state-of-the-art ensemble learning models in improving the performance and efficiency of IDS/IPS. We compared our approaches with other existing approaches using popular open-source datasets available in this area.
Jakka, Geethamanikanta and Alsmadi, Izzat M.
"Ensemble Models for Intrusion Detection System Classification,"
International Journal of Smart Sensor and Adhoc Network: Vol. 3:
2, Article 8.
Available at: https://www.interscience.in/ijssan/vol3/iss2/8