A COMPARATIVE STUDY OF BOOSTING CLASSIFICATION METHODS FOR DETECTION OF PARKINSON’S DISEASE

ABSTRACT

Parkinson's disease is a disorder of the nervous system that impacts the human brain. Several forms of research output are already in existence concerning Parkinson’s disease. Especially, its detection, prediction, diagnosis, and so on. This study aimed to compare the performance results of some boosting classification approaches for the detection of the same disease. For comparative analysis, three boosting classification methods include adaptive boosting, gradient boosting, and extreme gradient boosting were explored. The parameters of the classifiers, such as the learning rate, number of estimators, and random state, were altered for different testing, validation, and assessment purposes.

Results obtained showed that the eXtreme Gradient Boosting (XGBoost) method which is an optimized version of the boosting methods is capable of projecting the presence of Parkinson's disease with a high accuracy score compared to other boosting algorithms. It achieved approximately 98% accuracy, 97.5% precision, 100.0% recall, and 98.7% f1-score. Followed closely in performance by the gradient boosting method and then lastly adaptive boosting. Therefore, the high performance of XGBoost suggests its potential as a tool for early detection of Parkinson's disease, leading to improved patient outcomes.

Keywords: Boosting, classifier, methods, and Parkinson, comparison.