Data mining plays a crucial role in healthcare, combining elements of database management and statistical analysis to evaluate the effectiveness of medical treatments. Machine learning and data visualization techniques have become particularly valuable in this field. One significant health concern is heart disease linked to diabetes, a long-term condition that occurs when the body either does not produce sufficient insulin or is unable to use it effectively. Cardiovascular disease, a general term for conditions that involve the heart and blood vessels,is a major complication for diabetic individuals. While several classification algorithms are available for predicting heart disease, there remains a lack of sufficient data specifically focused on diabetic patients. Among various models tested, the decision tree approach consistently outperformed naive Bayes and support vector machines. By optimizing the decision tree, we improved its accuracy in assessing the risk of heart disease in diabetic individuals. This form of paper review considers and elaborates on the current use of ML in the earlier detection of various diseases on the basis of sound research. Firstly, the method of bibliometric analysis of the publication is implemented with scopus and WOS Data.