ParkDiag: A Tool to Predict Parkinson Disease using Data Mining Techniques from Voice Data

In the present decade there is more concentration on ageing disease. Some diseases are going to through their symptoms at early stage. The problem with aging disease like cancer, diabetes Alzheimer and Parkinson are if they are going to uncover at advanced stage it is very difficult to cure them totally , instated doctors can able to stop the growth of the disease up that stage where it was un covered. But it is highly difficult to them to cure the disease up to the root level. Parkinson disease is the one of the most painful and non-curable disease. It is going to occur from the age of 40 plus people. Few detailed clinico-pathological correlations of Parkinson's disease have been published. The pathological findings in 100 patients diagnosed prospectively by a group of consultant neurologists as having idiopathic Parkinson's disease are reported. There are some symptoms to identify the disease; the symptoms like to find the Parkinson Disease (PD). The dataset for the disease is collected from the people who are already having Parkinson. However, genetic causes of Parkinson’s are rare, only in approximately 6–8% of all cases. The details are collected from PD patients especially the voice data of PD patient were collated by recording their voice by pronouncing the ovals A,E,I,O,U repeatedly minimum 10 times form each PD patient. The results were analyzed by using data mining techniques, Bayes Net shows 70% , naïve bayes shows 83% , KStar and ADTree shows 100% and Random Forest classification results towards PD.