HEART DISEASE DIAGNOSIS USING MACHINE LEARNING CLASSIFICATION TECHNIQUES

Abstract

Predicting heart disease is difficult in medicine. In India, heart disease causes most deaths. In many nations, overwork, stress, and other factors cause cardiovascular disease deaths. It's linked to heart disease in adults. For identifying cardiac disease, a decision support system is needed. Our work uses data mining to better predict cardiac disease. Heart disease is a leading cause of mortality worldwide, notably in Bangladesh. Forecasting cardiac disease accurately is a difficult and time-consuming procedure, but machine learning (ML) methods may help. This article explains our preferred approach for predicting cardiac problems, which uses machine learning algorithms to discover key indicators and improve accuracy. The UCI Repository has 14 features from our dataset. We built our model by categorising the world using Maximum Entropy, Random Forest, and SVM. SVM delivered the best performance in our suggested system, with 92.67 percent accuracy for the threshold instances of the dataset. The new method has produced 20% more accurate results than before.