Scholarlink

Mobile devices have been essential to modern civilization over the last ten years, and they have played a direct role in defining the stages of development of mobile information access. As these contemporary mobile gadgets grow so quickly, security risks also increase rapidly, with malware being the most concerning of all. According to current research, Machine Learning is a useful and promising method for Detecting Malware on Android devices. To enhance performance, our approach integrates Multiple Machine Learning Techniques. We Proposed two techniques: 1. To reduce overall time, we deleted 14 highly correlated features as detected by Heatmap, resulting in an accuracy of 97.80161%. 2. To improve accuracy and reduce overall processing time, we employed a Confidence Score Technique, resulting in an accuracy of 98.14814%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Detection of Android Malware Through Diverse Machine Learning Techniques

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • C. Manikandan,
  • C. Veda Vardhan,
  • L. Sai Rishitha,
  • K. Saranya,
  • R. Amirtharajan,
  • A. Rajesh

摘要

Mobile devices have been essential to modern civilization over the last ten years, and they have played a direct role in defining the stages of development of mobile information access. As these contemporary mobile gadgets grow so quickly, security risks also increase rapidly, with malware being the most concerning of all. According to current research, Machine Learning is a useful and promising method for Detecting Malware on Android devices. To enhance performance, our approach integrates Multiple Machine Learning Techniques. We Proposed two techniques: 1. To reduce overall time, we deleted 14 highly correlated features as detected by Heatmap, resulting in an accuracy of 97.80161%. 2. To improve accuracy and reduce overall processing time, we employed a Confidence Score Technique, resulting in an accuracy of 98.14814%.