In the current era where data and information is the main fuel for business, education, banking, marketing and digital applications used by people. These applications consist of personal and sensitive data. These digital applications are using various technologies that makes it heterogeneous. This heterogeneity creates vulnerabilities in many cases and thus resulting in various threats like, virus, Trojan, spyware and other malicious programs. Malware is one of the deadliest threats in cyber security and it is very challenging to detect malware when entering a system because of its intelligent coding. The reverse engineering is used for detecting malware in classical way which is a time taking process and involves maximum human interventions. Therefore, there is also need of automated defensive mechanism to detect these malwares in order to reduce the duration of mitigation. The less time for accurate detection would result in stopping malware spreading in other part of the organization once it attacks the organization. Therefore, this paper focuses on the Machine Learning concept as one of the major solution for malware detection. This paper also addresses the classical reverse engineering concept that is used to analysis malware. The paper uses, Random Forest, XGBoosting Decision Tree, Gradient Boosting Classifier and SGD Classifier for analysis. Further the Machine Learning models that gives highest accuracy is XGBoosting that gives accuracy of 94.86% followed by random forest which 94.66%.