Parkinson’s disease (PD) is found to be a challenging issue which can offer a computerized estimate about classification of PD to patient people and healthy for normal people. Due to the importance of that problem, several types of biomedical data can be analyzed to
accurately detect PD by using different learning methods. This work considers the diagnosis of PD based on voice data by using non-linear support vector machine (SVM). However SVM is known as the one of the fast and accurate learning methods, selection of relevant feature elements of PD dataset can be effective on improving the classification performance of SVM. To this end, this paper proposed an SVM in parallel with GA based feature reduction model for selecting the most relevant features to get Parkinson's disease. The
GA-SVM resulted in improved accuracy, sensitivity and area under curve (95%, 98% and 92% respectively) compared to the other learning methods and feature selection algorithms. The GA-SVM provides a better, more accurate identification for presence of vocal disorder from speech recordings leading to more timely diagnosis.
Parkinson’s Disease (PD) Feature Selection Genetic Algorithm (GA) Classification Support Vector Machine (SVM)
Primary Language | English |
---|---|
Subjects | Software Engineering (Other) |
Journal Section | Articles |
Authors | |
Publication Date | June 1, 2020 |
Acceptance Date | March 15, 2020 |
Published in Issue | Year 2020 Volume: 3 Issue: 1 |
International Journal of Informatics and Applied Mathematics