Comparative Survey of Machine Learning Techniques for Prediction of Parkinson's Disease
Merry Saxena, Sachin Ahuja
Abstract
Prognosis and progression of Parkinson's disease is a critical question among the clinicians since there is a disparity of parameters taken into the diagnostic consideration thereby making the decision process difficult. Different datasets have been independently explored and applied through machine learning to analyze the incidence of occurrence and progression of the disease. The present paper is an updated report of the types of Supervised Machine Learning algorithms which have gained prominence within a span of last 5 years (2015- 2019). Further it highlights the use of hybrid intelligence models to improve the prediction accuracy and sensitivity over standalone methods. Conclusively the paper also emphasis on the need of development of multiparametric, big data based holistic predictive system.