The objective of the project titled "Malware Analysis and Detection Using Machine Learning Algorithm" is to improve the effectiveness of cyber-security measures by reliably detecting malicious software via the use of sophisticated machine learning protocols. In addition to being built using Python, the project makes use of the Flask web framework for backend operations. Additionally, HTML, CSS, and JavaScript are used in order to provide a frontend experience that is both responsive and interactive. The Extra Tree Classifier and Logistic Regression are the two machine learning models that are at the core of this research. A training accuracy of 97.42% and a testing accuracy of 97.23% are both achieved by the Extra Tree Classifier model, which displays exceptional performance. As a point of reference, the Logistic Regression model achieves an accuracy of 94.84% during training and 93.67% during testing. Both models are trained and verified with the use of the TUNADROMD dataset, which consists of 4465 cases and 242 characteristics. The target classification attribute is responsible for discriminating between malware and goodware.For the purpose of the study, a subset of 23 qualities was chosen and selected on the basis of their significance and influence on the categorization problem. With this deliberate choice, we hope to improve the performance of the model while simultaneously lowering the complexity of the computations. The findings of the experiment reveal that the Extra Tree Classifier is very efficient in discriminating between dangerous and benign software. As a consequence, it provides a trustworthy instrument for the detection of malware in applications that are used in the real world. In general, this study illustrates the effectiveness of machine learning algorithms in the field of cyber security. It offers a strong solution for the identification of malware that can be included into a variety of digital security infrastructures.