Recent years have seen a surge in malware attacks, which have caused considerable financial and operational damage to individuals and businesses alike. As malware techniques continue to evolve, it is essential to be proactive and adaptive in order to detect and reduce the risk of such attacks. To address this issue, this research is focused on creating a sophisticated malware detection system that utilizes machine learning algorithms to detect Malware attacks. With this technique, a comparative assessment of the algorithms used was carried out. The models were trained using four datasets. The machine learning algorithms used are Decision Trees, Xgboost, Logistic Regression, Naïve Bayes and Random Forest, are tested for their ability to detect Malware behaviors. A comparison of the results was carried using four evaluation metrics namely: Accuracy, Precision, Recall and F1-score. At the end of the experiments - For Dataset 1, XGBoost had the highest accuracy of 0.982456. For Dataset 2, Random Forest had the highest accuracy of 0.990432. For Dataset 3, XGBoost, Random Forest and Decision Tree had the highest accuracy of 1.0 meaning they achieved a perfect accuracy of 100%. For Dataset 4, both Random Forest and XGBoost had the highest accuracy with the values respectively 0.628474 and 0.628135. A bar chart was used to graphically represent our results for each evaluation metrics. Boxplot used to represent results showing the minimum, maximum, and quartile ranges.