One of the leading causes of death in the modern world is heart disease. One of the biggest problems in clinical data analysis is the prediction of cardiovascular disease. Making judgments and forecasts from the vast amounts of data generated by the healthcare sector has been demonstrated to be aided by machine learning (ML). Early detection of heart failure can be prevented with the accurate and prompt diagnosis of human heart disease, which also increases the prognosis of the patient. Manual methods to diagnose cardiac disease are prone to bias and interexaminer variability. Machine learning algorithms are effective and trustworthy tools for identifying and classifying heart disease patients and healthy individuals. According to various research, only a small portion of heart disease may be predicted using ML approaches. In this research, we suggest a unique approach that seeks to improve the accuracy of the prediction of cardiovascular illness by utilising machine learning approaches to identify key traits. We introduce SVM- with linear random forest for efficient prediction.