Optimizing Heart Disease Prediction Model with GridsearchCV for Hyperparameter Tuning

Forecasting heart disease is an imperative endeavor in healthcare, and accurate models are essential for early detection and intervention. In this proposed work, we have created a one-dimensional Convolutional artificial neural network model. with gridsearchCV hyperparameter tuning and compared the accuracy with a Utilizing Support Vector Machine (SVM) and a one-dimensional Convolutional Neural Network (1D CNN) with the Cleveland dataset. The SVM model is a widely used machine learning algorithm known for its effectiveness in classification tasks. We train the SVM model on the Cleveland dataset, which includes 13 features. The first 1D CNN model is trained on the same Cleveland dataset. It is designed to analyze one-dimensional data, in this case, the clinical and demographic attributes from the dataset. The second 1D CNN model incorporates hyperparameter tuning using GridsearchCV. This process investigates different combinations of hyperparameters to pinpoint the most effective setup for the model, enhancing its performance. To compare the accuracy of the three models, we evaluate their performance on a separate test dataset from the Cleveland dataset. By comparing the accuracies of the SVM model, 1D CNN model, and 1D CNN model with hyperparameter tuning, we gain insights into their predictive capabilities for heart disease. The results indicate that the SVM model achieves an accuracy of 81.96 percent, the 1D CNN model achieves an accuracy of 83.61 percent, and the 1D CNN model with hyperparameter tuning achieves an accuracy of 86.89 percent. These findings demonstrate the varying performance of the models in forecasting cardiovascular conditions using the Cleveland dataset. The findings offer insights into the efficacy of these models. and assist in selecting the most accurate Methodology for predicting heart disease.