AI-Based Android Malware Detection System Using Machine Learning For Mobile Security

Abstract

The rapid growth of Android applications has increased the risk of malware attacks on mobile devices, leading to threats such as data theft, unauthorized access, and privacy violations. Traditional signature-based malware detection techniques are often ineffective against newly emerging and evolving malware attacks. To address this issue, this project proposes an AI-Based Android Malware Detection System Using Machine Learning for Mobile Security that intelligently detects malicious Android applications using machine learning algorithms. The system is developed using the Django framework and integrates user authentication, dataset uploading, machine learning model training, and malware prediction functionalities. In this system, Android application datasets are uploaded and processed using the Pandas library. The dataset is divided into training and testing data using the train-test split method. A Support Vector Machine (SVM) classifier is used to train the model for identifying malicious and safe applications. The trained model is then stored using Joblib for future predictions. The proposed system allows authenticated users to upload test datasets and predict whether applications are malware or safe. The prediction results display the total number of applications analyzed along with malware and safe application counts. The system improves mobile security by providing fast, accurate, and automated malware detection. Experimental results demonstrate that the machine learning-based approach achieves high accuracy and effectively enhances Android malware detection compared to traditional methods. The project provides a scalable and user-friendly solution for securing Android devices against malicious applications and contributes to the advancement of intelligent mobile security systems using artificial intelligence and machine learning techniques.