Enhancing Cloud Security through Intrusion Detection: A Comprehensive Study Using the ISOT-CID Dataset
Year: 2024
Type: article
Abstract: Cloud computing has revolutionized data management, offering unparalleled scalability, flexibility, and efficiency. However, its open and multi-tenant nature introduces significant security vulnerabilities, making it an attractive target for cyber threats. Intrusion Detection Systems (IDS) tailored for cloud environments are essential in mitigating these risks. Despite various IDS models, benchmarking datasets representing realistic cloud environments is a substantial limitation. This study utilizes the ISOT Cloud Intrusion Detection Benchmark Dataset (ISOT-CID), a publicly available dataset featuring a range of network and application layer attacks collected from a real production cloud environment. The research explores the dataset's structure, analyzes attack patterns, and evaluates IDS models' performance to provide actionable insights for enhancing cloud security. This work contributes to the field by presenting a systematic analysis of ISOT-CID, identifying effective IDS models, and proposing improvements for future cloud intrusion detection research.
Author: Safana Alzide
Institution: Florida State College at Jacksonville
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4404011485