A Hybrid Intrusion Detection System for SDWSN using Random Forest (RF) Machine Learning Approach

Abstract

It is indeed an established fact which network security systems had certain technical problems that mostly tends to lead to security risks. Nowadays, Attackers could still continue to abuse the security vulnerabilities as well as shatter the systems and networks, and is quite pricey and even sometimes extremely difficult to resolve all layout and computing faults. The above appears to suggest that methodologies relying on preventive measures seem to be no longer secure and perhaps tracking of intrusion is necessary as a last line of defense. A Hybrid in Software Defined Wireless Sensor Network (SDWSN) the Intrusion Detection System is designed for this paper which really incorporates the benefits of Salp Swarm Optimization (SSO) algorithm as well as the classification of Machine Learning method it is based upon Random Forest (RF). We propose SSO optimization procedures to guarantee that the ideal features for the intrusion detector are chosen and in addition for improving the Random Forest (RF) classifier detection efficiency. To assess / calculate the reliability of the proposed approach here we make use of the generic NSL KDD dataset. Therefore, our proposed hybrid IDS-SSO-RF classifier further analyzes these detected abnormal activities. The known and unknown attacks are also identified. Hybrid framework also shown by the experimental results can reliably detect anomaly behavior and obtains better results in terms in terms of delay, delivery ratio, drop overhead, energy consumption and throughput.

Forecasting Carbon Dioxide Emission in Thailand Using Machine Learning Techniques

“…Accurately forecasting carbon dioxide (CO2) emissions has become a crucial problem as concern over climate change and its effects on the environment has grown. It has been demonstrated that machine learning [6], [7], [16]- [19] approaches are useful for modeling complex relationships and producing precise  ISSN: 2089-3272 IJEEI, Vol. 11, No.…”