Volume - 5 | Issue - 2 | june 2023
DOI
10.36548/jitdw.2023.2.004
Published
12 June, 2023
Heart failure affects a minimum of 26 million individuals and its occurrence has been increasing day-by-day. Heart failure occurs when the heart cannot pump enough blood to meet the body's needs. This is caused due to various reasons such as coronary heart disease, heart valve malfunctioning, diabetes, anaemia etc. So, it is important to predict heart failure in its early stage to reduce the mortality rate. Cardiovascular disease is the major contributing factor for the prediction of heart failure. This research uses Gaussian Naïve Bayes technique which comes under supervised learning algorithm to predict heart failure. Gaussian Naïve Bayes algorithm treats each feature to be independent and each feature has equal importance in predicting heart failure. For the purpose of predicting heart failure, the mean and standard deviation of each column in a dataset is used. Different datasets are used for the implementation of heart failure prediction, and performance metrics are identified for each dataset. The findings of this study suggest that the Gaussian Naive Bayes algorithm can be a useful algorithm in predicting heart failure, and it has the potential to improve patient outcomes by facilitating early detection of heart failure. There are various metrics used for the performance analysis of the system such as, accuracy, recall, precision, confusion matrix and ROC curve. Finally, this study concludes by presenting the implementation of heart failure using Gaussian Naïve Bayes.
KeywordsHeart Failure (HF) Supervised learning algorithms Gaussian Naïve Bayes (GNB) Performance metrics Receiver Operating Characteristic (ROC) Curve
