The Research On Parkinson Medication Recommended Model Based On Sacle Evaluation

Parkinson’s disease in the elderly is a common neurodegenerative disease, but with the diversity of people’s lives, middle-aged people suffering from Parkinson’s disease will be less. So far, there is no cure for Parkinson’s disease and is a more severe disease, but a good and necessary drugs surgical treatment can help Parkinson’s patients recover function. Currently, the diagnosis of Parkinson’s disease the most important basis of Parkinson’s disease results from the evaluation scale. Based on this study based on the evaluation scale automatic identification and treatment of Parkinson’s disease the recommended method for the prevention and treatment of Parkinson’s very important.This paper mainly through machine learning techniques to study recommendation model of Parkinson’s disease medication, thesis work are as follows:1) Proposed a support vector machine automatically identify Parkinson’s disease model, which is mainly based on a new scale designed to quickly identify Parkinson’s disease, and the new scale design is based on principal component analysis of Parkinson’s Disease Scale optimization algorithm. Through the experiment, we discover that the new combinations of scale which accounts for21%of the original western scales is highly comparable to original western scales for assessing Parkinson’s Disease using Support Vector Machine.2) Combines "k-labelset criterion" and "k nearest neighbor criterion," proposed hybrid strategy based on multi-mark learning Framework to be used for recommended for Parkinson’s disease drugs. The algorithm scales in Parkinson’s disease-the drug data sets than RAkEL, ML-kNN has a better performance.3) Based on the above work, design and implement based on "cloud platform+terminal" Parkinson polyclinics platforms, developed the server system and the Android platform.