Genetic Analysis with Feature Reduction to Predict the Onset of Parkinson’s disease
Ranjith N1, Lincy Mathews2

1Ranjith N, Department of Information Science and Engineering, Ramaiah Institute of Technology, Bangalore, India.
2Lincy Mathews, Assistant Professor, Department of Information Science and Engineering, Ramaiah Institute of Technology, Bangalore, India.
Manuscript received on July 13, 2020. | Revised Manuscript received on July 20, 2020. | Manuscript published on August 10, 2020. | PP: 12-16 | Volume-9 Issue-10, August 2020 | Retrieval Number: 100.1/ijitee.J73370891020 | DOI: 10.35940/ijitee.J7337.0891020
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Parkinson is a disease which directly affects the brain cells and certain movement, voice and other disabilities. Hence curable medication is not available in market. The best solution is the early diagnosis to relieve the symptoms of Parkinson’s disease affected people. One major concern effecting public is Parkinson’s disease (PD). This paper studies the bias of various traditional algorithms on the voice-based data that has various parameters recorded from Parkinson patients and healthy patients. A brief survey of techniques are mentioned for the prediction of Parkinson’s diseases is presented. To accomplish this task, identifying the best feature reduction approach was the primary focus. This paper further applies feature reduction techniques using a genetic algorithm for efficient prediction of Parkinson’s disease along with machine learning-based approaches. The proposed method also presents higher accuracy in prediction by using this optimal feature reduction technique. 
Keywords: Parkinson’s disease, Genetic algorithm, Feature reduction, Support Vector Machine, K-Nearest Neighbor, Naïve Bayes, Speech data.