Performance Analysis of Hybrid Cloud Intrusion Detection Model by Using Supervised Machine Learning Based Classification Algorithms
Year: 2025
Type: article
Abstract: Cloud Computing refers to an Internet-based infrastructure that delivers shared resources, software, and information to computers and other devices on an on-demand basis. However, it faces numerous security challenges, including issues related to availability, data confidentiality, integrity, and access control. Additionally, Cloud Computing is vulnerable to various conventional attacks. Traditional security systems are inadequate to protect Cloud services from these diverse threats. In the realm of Cloud Computing, Intrusion Detection refers to the identification and management of unauthorized access, harmful actions, and potential security threats. Intrusion Detection Systems (IDS) serve as security mechanisms that monitor network traffic and event logs to detect any anomalous activities. The Cloud Computing (CC) environment necessitates the implementation of specific Intrusion Detection Systems to safeguard each machine from potential attacks. Machine Learning and Deep Learning algorithms enhance the accuracy of Intrusion Detection Systems over time, leading to a decrease in both false positives and false negatives. In the hybrid phase, the implementation of Naïve Bayes and Decision Tree algorithms resulted in an impressive accuracy rate of 99.71%. Future research should explore additional combinations of hybrid models to achieve greater efficiency across all performance metrics.
DOI: 10.14419/mrn3dh08
Authors: Mr. D. Rajagopal, Dr. K. Padmanabhan
Institution: Periyar University
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4413442151