Predictive Modelling of Malware Behavior Using Machine Learning Algorithms

The research focuses on an innovative approach to use Random Forest, XG Boost and Decision Trees to enhance the malware prediction techniques. It helps to analyze large datasets containing behavior logs and relevant information. By leveraging the ensemble capabilities of Random Forest, XGBoost, and Decision Tree, the system can effectively capture intricate patterns and relationships within the malware behavior. Also, using Bagging Classifier enhances the robustness of the predictive model by aggregating multiple base learners and reducing over-fitting. The model outperforms other models by giving a high accuracy of 85 %. Thus, by combining the strengths of machine learning models, the system can detect anomalies and identify potential malicious activity, thereby strengthening cybersecurity defenses.