Answers are grounded in this paper and cite their sources.
Malware detection has become increasingly challenging due to the rapid evolution of sophisticated cyber threats and the limitations of traditional detection techniques. Conventional approaches such as signature-based and heuristic methods are ineffective against zero-day and polymorphic malware, while machine learning models rely heavily on manual feature engineering. Deep learning techniques have improved detection accuracy; however, they often lack interpretability, which is critical for cybersecurity applications. This paper proposes an AI-driven malware detection framework based on a hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model integrated with explainable artificial intelligence techniques. The proposed model leverages CNN for spatial feature extraction and LSTM for capturing temporal dependencies in malware behavior, enabling comprehensive analysis of both static and dynamic characteristics. To enhance model transparency, SHAP and LIME are incorporated to provide global and local explanations of predictions. The model is evaluated using benchmark datasets and standard performance metrics, including accuracy, precision, recall, and F1-score. Experimental results demonstrate that the proposed CNN–LSTM model achieves an accuracy of 96.8%, outperforming traditional machine learning models and standalone deep learning approaches. The explainability analysis further confirms that the model relies on meaningful features such as entropy and API call patterns, improving trust and interpretability. The proposed framework provides a robust, accurate, and interpretable solution for malware detection, contributing to the development of trustworthy AI-based cybersecurity systems.
No takes yet. Share an insight, caveat, or question.
Km. Muskan (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: