Scholarlink

In the current digital age, Android malware attacks have become a critical concern, as they directly threaten the security and privacy of user’s personal information. Traditional malware detection approaches rely on signature-based techniques, and static analysis often falls short in detecting evolving malware. Therefore, innovative detection techniques are essential to safeguard user data. This study proposes a convolutional neural network (CNN)-based Android malware classification framework using image data from 25 malware families. Experimental results demonstrate the effectiveness of our image-based detection framework, achieving 91.33% accuracy with an impressively lower model loss of 0.27. This research underscores the promise of image data analysis as a transformative approach to improving Android malware detection, setting the stage for further innovations in cybersecurity.

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

Advancing Android Malware Detection: CNN-Based Image Analysis Framework

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Diptimayee Sahu,
  • Satya Narayan Tripathy

摘要

In the current digital age, Android malware attacks have become a critical concern, as they directly threaten the security and privacy of user’s personal information. Traditional malware detection approaches rely on signature-based techniques, and static analysis often falls short in detecting evolving malware. Therefore, innovative detection techniques are essential to safeguard user data. This study proposes a convolutional neural network (CNN)-based Android malware classification framework using image data from 25 malware families. Experimental results demonstrate the effectiveness of our image-based detection framework, achieving 91.33% accuracy with an impressively lower model loss of 0.27. This research underscores the promise of image data analysis as a transformative approach to improving Android malware detection, setting the stage for further innovations in cybersecurity.