HeartInspect: Heart Disease Prediction of an Individual Using Naïve Bayes Algorithm

Heart disease is a serious health issue that contributes significantly to the high death worldwide. Therefore, the creation of a reliable system for heart disease prediction is essential for early intervention and better results. Such technologies can help identify at-risk persons and enable prompt preventive interventions by utilizing cutting-edge algorithms and analysing pertinent data. Nonetheless, predicting heart disease is a difficult endeavour, especially in underdeveloped regions with few diagnostic tools and a shortage of trained medical workers. Moreover, the healthcare sector produces a tremendous quantity of data on cardiac disease, yet these important resources are frequently underutilized when it comes to making well-informed decisions. Additionally, pricey heart diagnostics like electrocardiograms are now out of reach for the typical person due to increased living expenses. This present work suggests a Naive Bayes-based cardiac disease prediction system as a solution to these problems. The system makes use of a dataset that includes information about a person's heart disease status, height, weight, physical health, difficulties walking, age group, physical activity, general health, and sleep duration. For training and testing purposes, the dataset is partitioned 80/20. The dataset is analysed using the Naive Bayes technique to determine the chance of cardiac disease. Despite assuming independence among the characteristics, the system shows promising performance, reaching about 71–73% precision in heart disease prediction. Even though this falls short of higher standards, it is nevertheless a noteworthy accomplishment in light of the difficulties mentioned in the problem statements. In summary, the Naive Bayes algorithm-based heart disease prediction system reported in this work shows promise for predictions that are 71–73% accurate. This method helps address the problems involved with cardiac disease prediction, particularly in environments with limited resources, by making use of the data that is currently available and overcoming resource constraints.