Android Malware Detection and Prevention Based on Deep Learning and Tweets Analysis

Smartphones play a pivotal role in human life, underscoring the critical importance of security and privacy. This is particularly true for the Android operating system, which dominates the smartphone market. However, its widespread usage makes it a prime target for malware developers, posing significant risks, especially to Internet of Things (IoT) devices relying on Android applications. This paper introduces a multi-layer method for Android OS malware detection. Our approach uniquely integrates real-time data extraction from Twitter and deep learning techniques. We utilize Twitter data to update our malware hash database every 48 hours, effectively capturing the latest malware signatures. Additionally, we employ a deep learning model based on a Recurrent Neural Network (RNN) with Long Short-Term Memory (LSTM) to analyze application permissions, achieving a 94% accuracy in malware detection. This multi-layer system provides a comprehensive solution to Android OS security challenges