Heart Disease Prediction with Decision Tree

Heart disease remains a major global health issue, which emphasizes the need for accurate prediction models to aid early diagnosis and effective intervention. This study explores the use of a Decision Tree algorithm to predict heart disease. The dataset used consists of 272 entries, with seven key variables including age, sex, blood pressure, cholesterol, fasting blood sugar, maximum heart rate, and ST depression. The data underwent preprocessing to handle missing values and convert categorical data into numerical format. The model was trained with 80% of the data and tested with the remaining 20%. Evaluation metrics such as accuracy, precision, recall, and f1-score, were used to evaluate the model’s performance. The results demonstrated the model’s efficacy, achieving an accuracy of 81.48%, a recall of 82.93%, a precision of 91.89%, and an f1-score of 87.18%. These results highlight the potential of the Decision Tree Algorithm in heart disease prediction, particularly for its simplicity and interpretability. Despite the study’s limitations, such as the small dataset size that could affect generalizability, this study demonstrates significant predictive potential and a strong foundation for future work. Future research should explore alternative machine learning algorithms to improve prediction accuracy and enhance the model’s robustness for real-world application.