CADS-ML/DL: efficient cloud-based multi-attack detection system

With the increasing adoption of cloud computing, securing cloud-based systems and applications has become a critical concern for almost every organization. Traditional security approaches such as signature-based and rule-based have limited detection capabilities toward new and sophisticated attacks....

Full description

Saved in:
Bibliographic Details
Published in:International journal of information security 2023-12, Vol.22 (6), p.1989-2013
Main Authors: Farhat, Saida, Abdelkader, Manel, Meddeb-Makhlouf, Amel, Zarai, Faouzi
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Artificial intelligence
Cloud computing
Coding and Information Theory
Communications Engineering
Computer Communication Networks
Computer Science
Cryptology
Cybersecurity
Datasets
Decision trees
Deep learning
Effectiveness
Intrusion detection systems
Machine learning
Management of Computing and Information Systems
Networks
Operating Systems
Regular Contribution
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the increasing adoption of cloud computing, securing cloud-based systems and applications has become a critical concern for almost every organization. Traditional security approaches such as signature-based and rule-based have limited detection capabilities toward new and sophisticated attacks. To address this issue, there has been an increasing focus on implementing Artificial Intelligence (AI) in cloud security measures. In this research article, we present CADS-ML/DL, an efficient cloud-based multi-attack detection system. We investigate the effectiveness of Machine Learning (ML) and Deep Learning (DL) techniques for detecting cloud attacks. Our approach leverages a realistic dataset consisting of both benign and fourteen common attack network flows that meet real-world criteria on the AWS cloud platform. We evaluate eight Intrusion Detection Systems (IDSs) based on ML and DL algorithms, including Decision Tree (DT), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Gated Recurrent Units (GRU), Long Short-Term Memory (LSTM), Stacked LSTM, and Bidirectional LSTM (Bi-LSTM) models. Experimental results demonstrate that the CADS-ML/DL system, specifically the XGBoost model, outperforms the other models, exhibiting an accuracy of 0.9770 and a false error rate of 0.0230. Furthermore, we validate the effectiveness of our proposed XGBoost model on the AWS benchmark CSE-CICIDS2018 dataset, attaining a remarkable accuracy score of 0.9999 and an exceptionally low false error rate of 0.0001. Our findings suggest that AI-based approaches have the potential to detect cloud attacks effectively and contribute to the development of reliable and efficient IDSs for cloud security.
ISSN:1615-5262
1615-5270
DOI:10.1007/s10207-023-00729-4