Research on novel ways to detect malwares through data collection, reverse engineering, data pre-processing, feature selection, machine learning algorithm training, and model evaluation.
Collect a dataset of clone apps from various sources, including third�party app stores, forums, and websites. APK to Smali Conversion: Use Kali Linux tools such as apktool and dex2jar to convert the APK files of the clone apps into Smali files.
Extract relevant data such as dalvik opcodes and frequently used malicious keywords. Clean and format the data to remove irrelevant or redundant information.
Apply feature selection techniques such as chi-squared, mutual information, and correlation-based feature selection to select the most important features for use in the machine learning algorithms. Algorithm Training: Train decision tree algorithms such as C4.5, ID3, and CART using labeled data to create a decision tree model. Evaluate the performance of the model using performance metrics such as accuracy, precision, and recall.
Deploy the final model to enhance mobile device security by detecting and preventing the installation of malicious clone apps. Evaluation: Evaluate the performance of the deployed model using a testing dataset and refine the model if necessary. By using Kali Linux tools such as apktool and dex2jar, we can generate Smali files from APK files, which will be used in the pre-processing step to extract relevant 7 data for feature selection and algorithm training. This updated workflow utilizes data collection, APK to Smali conversion, pre-processing, feature selection, algorithm training, model development, model deployment, and evaluation to enhance mobile device security by detecting and preventing the installation of malicious clone apps.