Abstract

Parkinson’s Disease (PD) is a neurodegenerative disorder that affects millions of people worldwide. It is a progressive disorder that results in numerous complications. Early detection of the disorder is essential for effective management and treatment of PD. Used two different methods for PD classification using vocal biomarkers. First, we used LASSO-based feature selection using SMOTE oversampling techniques. Second, we used traditional machine learning algorithms along with comprehensive feature analysis. The proposed system is based on 195 samples and 23 vocal features such as jitter, shimmer, and other vocal characteristics. Experimental results showed that Random Forest achieved 94.87% accuracy along with 96.97% F1-score using vocal biomarkers. The current study used six different machine learning algorithms for PD classification using vocal biomarkers. These algorithms include logistic regression, decision tree, random forest, support vector machines, K-NN, and naive bayes. The experimental results showed that the proposed ensemble methods, such as random forest and decision tree algorithms, resulted in high accuracy compared to other algorithms. These algorithms resulted in 94.87% and 92.31% accuracy in classifying PD by utilizing vocal features. LASSO regularization and correlation analysis were utilized in the proposed system to select features for the detection of Parkinson’s disease. The study compares six machine learning algorithms: Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Naive Bayes. Results indicate that ensemble methods, particularly Random Forest and Decision Tree, outperform other approaches with accuracies of 94.87% and 92.31% respectively. Feature selection using LASSO regularization and correlation analysis identified key vocal biomarkers for PD detection.

Authors

V. Indumathi
RVS College of Arts and Science, India

Keywords

Parkinson’s disease, Machine Learning, Vocal Features, LASSO Regression, Feature Selection, Classification, Random Forest, SMOTE, Acoustic Analysis