Improving Parkinson's Disease Prediction Accuracy: A Comparative Analysis of Logistic Regression and Decision Tree Models

This study aims to determine the effectiveness of predicting Parkinson's Disease and to evaluate the accuracy of different predictive models. Specifically, it compares the performance of Logistic Regression against Decision Tree algorithms in identifying whether an individual has Parkinson's Disease. The research involves analyzing each model's predictive accuracy with sample groups of 10 individuals, utilizing parameters such as an alpha value of 0.05, a beta value of 0.2, and a power value of 0.8. Our findings indicate that Logistic Regression achieves a higher accuracy rate of 93.95% compared to the Decision Tree's 89.03%. This substantial difference demonstrates that Logistic Regression is notably more effective at predicting Parkinson's Disease. The Logistic Regression model shows a clear advantage over the Decision Tree model, making it a more reliable choice for predicting the disease. This result highlights Logistic Regression as a superior method for accurate Parkinson's Disease prediction, offering a promising approach for future diagnostic practices and research in this area.