A Systematic Approach to Detect Parkinson’s Disease using Traditional and Ensemble Machine Learning Techniques

Abstract

Parkinson's disease (PD) is a neurodegenerative condition that worsens with time and affects both the neurological system and body components under the control of the nervous system. This condition causes slow movements, tremors, balance problems, and more. Currently, we have no proper cure or treatment available, but it can sometimes be cured with medication if it is diagnosed in its initial stages. Voice deterioration is also a common symptom, which often presents in the initial stages of the disease. As a result, the project 'A Systematic Approach to Detect Parkinson’s Disease Using Traditional and Ensemble Machine Learning Techniques' is used to detect PD using voice data. In order to create a model that is capable of accurately identifying the disease's existence in a person's body, this project makes use of a variety of machine learning techniques, ensemble learning approaches, and Python libraries. This work aims to compare various machine learning models in the successful prediction of PD and develop an effective and accurate model to help detect the disease at an earlier stage, which could help doctors assist in the cure and recovery of PD patients. This project showed 97% efficiency. For this purpose, we plan to use the Parkinson’s disease dataset in [5] , which is acquired from the UCIML repository.