A Comprehensive Exploration of Machine Learning and Explainable AI Techniques for Malware Classification

The rapid evolution of malware presents a significant challenge to cybersecurity, necessitating more effective detection and classification methods. Malware attack is a vulnerable and prevalent threat in cyber security. Awareness about these threats and regular updation of security measures are critical. There are various techniques in which we can detect the malwares and prevent them from occurring. Identification of malware is crucial in this technological era to prevent destruction. In this work, we have used various machine learning classifiers including ensemble techniques such as boosting and bagging classifiers to classify four different types of Android Malwares. Random Forest has outperformed with a F1-score of 88%, and model Explainability has been deployed to reveal the most significant feature.