Sign in Subscribe
Machine learning

Novel Wavelet-based Model for Android Malware Detection

Strategic Insights

— 1 min read

A team of researchers from Abdelmalek Essaadi University has developed a novel approach for detecting Android malware using a wavelet-based model that enhances system call feature selection. The model leverages the Chi-Square test for feature selection and utilizes the Haar wavelet to transform the selected attributes into wavelet coefficients. The researchers evaluated several machine learning classifiers, including Decision Tree, Support Vector Machine, Random Forest, and Neural Network, and found that the Random Forest classifier, combined with wavelet-based feature selection, achieved the highest performance, attaining an impressive accuracy rate of 99.99%.

Key Takeaways:

  • The proposed wavelet-based model for Android malware detection utilizes system calls features, which are extensively used in machine learning methods for malware detection.
  • The model employs the Chi-Square test for feature selection and utilizes the Haar wavelet to transform the selected attributes into wavelet coefficients.
  • Several machine learning classifiers, including Decision Tree, Support Vector Machine, Random Forest, and Neural Network, were evaluated in the experiments.
  • The Random Forest classifier, combined with wavelet-based feature selection, achieved the highest performance, with an accuracy rate of 99.99%.
  • The model was evaluated using metrics such as Accuracy, Recall, F-Score, and Precision.
  • The research concludes that the proposed model is effective in detecting Android malware, with results unequivocally demonstrating its effectiveness.

Statistics:

  • Accuracy: 99.99%
  • Recall: Not specified
  • F-Score: Not specified
  • Precision: Not specified
  • Number of machine learning classifiers evaluated: 4 (Decision Tree, Support Vector Machine, Random Forest, and Neural Network)
  • Number of Android malware types detected: 4 (adware, riskware, banking malware, and SMS malware)
  • Number of system calls features used: Not specified
  • Number of wavelets used: 1 (Haar wavelet)
  • Number of machines used in the experiments: Not specified

Sources:

  • A Novel Wavelet-based Model for Android Malware Detection Utilizing System Calls Features. Journal of Network and Systems Management, 2025;33(3). (Springer - www.springer.com; Journal of Network and Systems Management - www.springerlink.com/content/1064-7570/)
  • NewsRx. Abdelmalek Essaadi University Details Findings in Machine Learning (A Novel Wavelet-based Model for Android Malware Detection Utilizing System Calls Features). Information Technology Newsweekly. July 8, 2025; p 67.

Read more