Heart disease, also called cardiovascular disease, is considered one of the deadliest diseases that cause high mortality worldwide. Early detection or prediction is a challenging task in the medical field. There is a massive amount of data in the healthcare industry, and processing this amount of data is a tedious task. A computer-aided system that predicts cardiac disease can save time and money. Researchers have researched several computer-assisted diagnoses for disease prediction and prognosis. In this paper, the authors provide an extensive literature survey of various classification approaches such as Machine Learning, Feature Selection, Hybrid, Ensemble, and Deep Learning used by researchers in the last decade for Heart Disease prediction. Furthermore, as the paper focuses on Machine Learning techniques, comparative analysis of the performance and accuracy of various Machine Learning techniques are summarized in tabular form. Additionally, this work critically assesses earlier methods and outlines their shortcomings. Finally, the article offers some potential future research direction in machine learning-based automated heart disease prediction.