Enhanced Heart Disease Prediction Using Grid Search Optimized XGBoost Classifier

Since heart disease is still one of the major causes of death worldwide, there has been a lot of study done on early diagnosis and prevention techniques. Thus, an XGBoost classifier tuned for Grid Search is used to carry out a novel prediction. Finding the best characteristics in the dataset is aided by Grid Search optimization. The preprocessed dataset is subjected to suitable feature selection procedures in order to extract the most pertinent predictors. The accuracy, precision, recall, and F1-score are among the performance indicators used to assess the suggested method. Comparing the suggested model to other techniques as the Random Forest model, Decision Tree model, and Multi-Layer Perceptron (MLP), it yielded an accuracy of 97.56% and a Fi_Score of 97.69%. By enabling early identification and preventive actions, this research adds to the ongoing efforts to develop accurate and dependable predictive models for cardiac disease, which in turn improves patient outcomes and lessens the strain on healthcare systems.