Android-IoT Malware Classification and Detection Approach Using Deep URL Features Analysis

Currently, malware attacks pose a high risk to compromise the security of Android-IoT apps. These threats have potential steal critical information, causing economic, social, and financial harm. Because their constant availability on network, Android apps are easily attacked by URL-based traffic. In this paper, an classification detection approach using deep broad URL feature mining is proposed. This study entails development novel traffic data preprocessing transformation method that can detect malicious network analysis. The encrypted mined decrypt transmitted data. To extract sequenced features, N-gram analysis used, afterward, singular value decomposition (SVD) utilized reduce features while preserving actual semantics. latent extracted semantic tool. Finally, CNN-LSTM, multi-view learning approach, designed for effective detection.