Cardiovascular disease has become a primary reason for death globally, and predicting its occurrence is a crucial task in the medical field of data analytics. Machine learning has demonstrated its value in the healthcare sector by examining a large amount of data. However, there are limited studies on using ML techniques to predict heart disease accurately. This study introduces an approach that utilizes machine learning techniques to identify important features and enhance cardiovascular disease prediction. The forecasting model proposed in this study utilizes various combinations of features and classification techniques, resulting in a precision level of 88%. The model employs a hybrid version of a random forest with the linear model to achieve this level of accuracy.