Ensemble Learning Approaches for Detecting Parkinson's Disease

Parkinson's disease is a neurodegenerative disorder that occurs in elder people and affects movement with visible symptoms gradually escalates to a maximum over a period of time. Basic body functions namely walking, hearing, speaking, etc., are affected by this disease. Analysis of this disease can be done using ensemble learning algorithms that produce good results. As a result, the best one picked will have the maximum accuracy in determining if the patient has the condition. Dataset is obtained from the UCI ML (Machine Learning) depository, and is named Parkinson disease dataset which has repeated features that are acoustic in nature and contains a list of 240 cases with 48 different features whose performance metrics are measured by utilizing various ensemble learning techniques. As a consequence, the ideal outcome is chosen with the greatest precision since applications in medical management often demand greater precision and efficiency is of the utmost importance. Random forest, Bagging, AdaBoosting and Gradient Boosting are the models used in the process. These models can be useful to doctors in predicting disease by anticipating the symptoms exhibited in patients.