To search, Click below search items.


All Published Papers Search Service


Evaluation of Supervised Learning Model for Heart Disease Diagnosis


Annapurna. H. S, Tsehay Admassu Assegie, Sushma S.J, Padmashree S and Bhavya B.G


Vol. 22  No. 11  pp. 553-556


This research evaluates the performance of the supervised learning method for heart disease prediction. The study compares the result of different supervised learning methods with performance measures such as accuracy, recall, precision, F-score, and receiver operating characteristic curve (ROC) on the Cleveland heart disease dataset collected from the publically available Kaggle data repository. The dataset is pre-processed using a min-max scaler before the supervised learning model is trained. The result reveals that the decision tree, random forest, and gradient boosting scored ROC Area under a Curve of 1.00. Moreover, the decision tree, random forest, and nearest neighbor score 1.00 precision. Support vector machine, linear discriminant analysis, and the Na?ve Bayes have low performance compared to the neural network, nearest neighbor, and AdaBoost model.


Supervised learning, Cleveland, Kaggle, prediction, health informatics.