,
Deputy Director General, Scientist-G, Department of Computer Science and Engineering, National Informatics Centre (NIC), Koneru Lakshmaiah Education Foundation , Guntur, Andra Pradesh , India
Assistant Professor, Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation , Guntur, Andhra Pradesh , India
Cloud computing is a critical infrastructure to the modern digital services, which provides the ability to store data on a scale, distributed computing, and the ability to deploy services flexibly. Moreover, the high rate of cloud environment development has also contributed to the risk of malicious intrusions like the spread of malware, unauthorized access, insider threats, and suspicious network activity. Such threats are hard to detect because of the very high dimensionality of cloud activity datasets and redundant or irrelevant attributes. This research suggests a Dynamic Attribute Filtration framework to identify malicious activities in cloud environments with high accuracy to report this issue. The proposed system dynamically determines the importance of attributes based on statistical measures of importance (information gain and correlation analysis), and selects the useful features based on an adaptive threshold mechanism. The filtered feature set is then used by a machine learning classifier to differentiate between normal and malicious cloud activities. It was tested with Python and traditional cloud security datasets with thousands of networks and system activity records. According to the Investigational results, the proposed method considerably extends detection performance in opposition to the traditional feature selection methods. The explicit model has an accuracy of 98.2%, precision of 97.8%, recall of 98.5%, and a F1-score of 98.1% with a false positive rate of 1.6%. The comparative analysis, with no filtering and all feature models, had an accuracy of 94.1%, and the static feature selection methods led to an accuracy of about 95.6. The proposed framework saved the time of computational processing approximately 20-25%, which is more efficient when it comes to large-scale data analysis of clouds. The findings indicate the effectiveness of dynamic attribute filtering in developing malicious activity recognition in cloud settings. The proposed framework increases the detection accuracy, minimizes false alarms, and provides an efficient method to protect modern cloud infrastructures.
This is an open access article distributed under the Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.
