The healthcare industry produces a lot of data most of which can be useful for decision making especially on heart related diseases and in the global healthcare industry, heart disease is considered to be a major cause of deaths. Current diagnostic procedures are also not very efficient and accurate due to the time-consuming nature of manual data analysis techniques. To solve these problems, we introduce an Enhanced Linear Support Vector Machine (ELSVM) algorithm that can be used for detecting and avoiding important factors that are directly linked to heart disease. Preprocessing includes removing outliers through the Interquartile Range of Data Normalization (IQDN) and feature selecting through Recursive Feature Elimination (RFE) in order to exclude features that are insignificant. The ELSVM method combined with the Machine Learning (ML) approach focuses on improving the chances of early detection and prevention of heart diseases. This approach aims to enhance sensitivity, specificity and the time dimension and to offer a reliable instrument for the forecasting cardiovascular disease impact and the enhancement of diagnosis.