PREDICTION OF HEART DISEASE USING HYPER PARAMETER TUNING ON RANDOM FOREST CLASSIFIER

Machine learning algorithms are frequently used in the healthcare business to anticipate fatal illnesses. The purpose of this research was to compare the performance of a conventional system with a suggested system that predicts cardiac illness using a Random Forest classification model. Using the grid search technique to the Random Forest classification algorithm, the proposed system assisted in tuning the hyperparameters. The performance of the heart disease prediction system is the major focus of the research. The hyperparameter tweaking model may be used to boost the performance of prediction models. The classic and new approaches were evaluated in terms of accuracy, precision, recall, and F1 score. The proposed hyperparameter tuning model achieved accuracies in the range of 85.25 percent to 91.80 percent, The old approach has accuracies ranging from 81.97 percent to 90.16 percent. When compared to the old technique, these tests demonstrated that the suggested prediction approach is capable of delivering more accurate findings in predicting heart disease with the acquisition of achievable performance.