Parkinson disease is becoming more prevalent, in the ageing population, making it difficult to comprehend the disease mech- anisms. ML provides a promising result in identifying symp- toms, which enables the accurate identification of persons af- fected by parkinson disease. This study aimed to develop various machine learning algorithms with automatic feature selection ca- pabilities for detection. Here is a selection of five machine learn- ing approaches to identify parkinson disease Linear discriminant Analysis, KNN, Logistic Regression, Random Forest and XG Boost, ensuring robust and unbiased results. The performance metrics used to evaluate the models included classification accu- racy, recall, precision, and F1-score. In this analysis, a compari- son of all algorithms has been included and observed that the KNN algorithm yielded the best promising results to detect par- kinson disease with 94.87% classification accuracy, with recall, precision, and F1-score all more than 94%. These results suggest that ML-based PD detection systems have the potential to im- prove early diagnosis, leading to better patient outcomes and re- duced healthcare costs. Consistent with Sustainable Development Goal 3, which en- deavours to "ensure healthy lives and promote well-being for all at all ages," the results of the study have the potential to enhance the quality of life for those impacted by parkinson disease.