Traditional antivirus methods have proven insufficient

The gap

Evidence profile

Sourced from the stated research gap and future work of the source papers, classified as general, spanning 3 journals.

Supporting evidence — 3 representative gaps

• Intelligent malware detection on Android smartphones via a hybrid approach using gradient boosting and convolutional neural network (2026) · Scientific Reports · doi

  Existing techniques demand extensive feature engineering and representation, leading to higher computation times and error rates. The lack of a robust and effective hybrid approach that combines the strengths of Convolutional neural networks and Gradient Boosted Machines for Android malware detection. The need for a technique that can achieve notable improvements in accuracy, precision, recall, and AUC, and significant reductions in false positive rate and error rate.

  generalstated research gap
  Keywords: existing techniques demand extensive feature engineering representation leading
• Deep Learning Ensemble Model for Precise and Robust Android Malware Detection (2026) · International Journal of Drug Delivery Technology · doi

  Android's adaptability and ease of use have caused it to swiftly rise to the top of the mobile operating system market. The vast majority of persistent malevolent attacks also target it. This necessitates the quick installation of a strong malware detection system. With an accuracy of 99.16%, precision of 99.98%, recall of 99.87%, and F1score of 99.76%, this study's CNN+RNN hybrid model for Android malware classification achieves impressive results on the CIC-InvesAndMal2019 dataset. The suggested model proves in cybersecurity by surpassing more traditional ML algorithms like DT, RF, and SVM. However, challenges such as overfitting, indicated by fluctuations in validation loss, and high computational demands during model training, remain. To address these limitations, future work should focus on enhancing model regularization techniques, optimizing training procedures for faster convergence, and utilizing transfer learning to improve scalability.

  generalfuture workevidence 5/5
  Keywords: model android system malware training adaptability ease caused swiftly rise mobile operating market vast majority
• DeepTrust: Multi-step classification through dissimilar adversarial representations for robust android malware detection (2026) · Expert Systems with Applications · doi

  Traditional antivirus methods have proven insufficient in safeguarding mobile users, especially against encrypted and zero-day malware. Machine learning-based approaches are vulnerable to adversarial examples. There is a need for a robust Android malware detection system that can withstand feature-space attacks.

  generalstated research gapevidence 5/5
  Keywords: traditional antivirus methods have proven insufficient safeguarding mobile

Questions about this gap