An Intrusion Detection System for Preserving Information Security in Cloud Environment

Abstract

Cloud computing acquires additional security risks needing Intrusion Detection Systems (IDS) which aid in confirming the safety and dependability of systems and networks in the cloud environment. Increasing cloud security is the focus of this paper which proposes a framework of The Probabilistic Optimized Feature Voting Classification (POFVC) which aims to strengthen intrusion detection systems mechanisms in cloud computing environments. In this regard, the paper proposed the POFVC model which employs probabilistic optimization and feature voting to enhance the precision and production of intrusion detection. POFVC integrates machine learning techniques to fight the feature selection challenge at a progressive level, showing great outcomes on datasets widely known for intrusion detection such as UNSW-NB15, KDD Cup 1999, NSL-KDD, CICIDS2017, AWID, and DARPA Intrusion Detection Data. With accuracy greater than 98%, POFVC outshines conventional models SVM and Logistics, as well as other further evaluation metrics.