Abstract:
An ever-increasing number of malicious software programs are creating code to attack vulnerabilities in Android Machines due to the widespread adoption of Android-based d...Show MoreMetadata
Abstract:
An ever-increasing number of malicious software programs are creating code to attack vulnerabilities in Android Machines due to the widespread adoption of Android-based devices among consumers. It is anticipated that this growth in the usage of Android machines will move in the forward direction at a high rate. An adversary may easily get into machines that are running the Android operating system since there are a large number of available security apps, many of which are not allowed, and there is a substantial risk that sensitive data will be compromised. Using ensemble learning, SmRM, the Ensemble Learning is a cutting-edge riskware identification and mitigation methodology that is proposed in this research paper. To identify and get rid of riskware risks, SmRM Ensemble Learning employs machine learning, behavioural analysis, and signature-based techniques. In this research work, the idea proposed research methodology employ application to early detect riskware attacks on android machine by getting intelligence from a machine learning-trained model to determine whether or not the data packet is malicious. The results prove that ensemble learning with maximum depth 4, gets improved results in terms of ROC Curve, recall, precision and F1 score for identifying attack packets and benign packets through traffic patterns.
Published in: 2023 Annual International Conference on Emerging Research Areas: International Conference on Intelligent Systems (AICERA/ICIS)
Date of Conference: 16-18 November 2023
Date Added to IEEE Xplore: 13 February 2024
ISBN Information: