Parkinson’s Disease (PD) is a neuro-degenerative disorder that affects the motor skills of a person when the production of dopamine is reduced and eventually motor dysfunction. Dopamine is a chemical produced by brain cells, which is responsible for controlling movements. Thus, if Parkinson’s disease is detected earlier, then the progression and the effects of its symptoms can be controlled. The proposed Parkinson’s Disease Detection (PDD) model used for the detection of Parkinson’s Disease using various Machine Learning (ML) algorithms such as K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Logistic Regression (LR), Voting Classifier, and Random Forest (RF). The algorithms are applied on balanced data scales and on features selected using LDA and PCA from the speech recording dataset. On experimental analysis, the SVM algorithm is found to be more efficient on the balanced dataset and PCA extracted features, giving an accuracy of 96.73% and 95.42% respectively as compared to the other classification algorithms. While the Random Forest algorithm gives the best results on LDA applied dataset giving an accuracy of 95.85%.