For the past two decades, heart diseases have been the major cause of deaths worldwide. Various risk factors along with aging population, and evolving lifestyles have been major contributors for their prevalence. The World Health Organization has made various efforts to emphasize the need to address the ongoing challenge of cardiovascular diseases on a global scale. Adopting Machine Learning (ML) algorithms can be utilized by health care officials to study and analyze the risk of various medical conditions and furthermore, come up with better diagnostic tools to treat the same. The Machine Learning tools study the datasets provided to them, analyse them for patterns and scan the factors for relationships among them gaining knowledge on how every factor affects the final outcome. The primary focus of this research is based on the implementation of advanced machine learning techniques to predict and prevent heart diseases using the Cleveland dataset. Careful preprocessing which includes outlier removal, missing data imputation and feature scaling has been applied on the dataset. Exploratory data analysis has been conducted to understand the relationships between various factors contributing to heart diseases. Amongst the various ML techniques used, XGBoost has shown impressive accuracies of up to 95%. In addition to that, the ROC curve analysis exhibits an exceptional under the curve area of 0.99 emphasizing on the model’s ability to differentiate between positive and negative outcomes efficiently. Findings from this study underscore the efficacy of employing ML techniques. The results obtained help contribute to the ongoing efforts of leveraging data driven methods to enhance the prediction and prevention of cardiovascular diseases.