Highlights: ? Haar wavelets is used first time with ensemble learning. ? Results obtained is far better than existing results. ? Robust inference system is proposed for Parkinson’s disease for the first time. ? Sparse Multinomial Logistic Regression is given due importance in solving Parkinson’s problem. ? KDE used first time.Abstract: This work presents more precise computational methods for improving the diagnosis of Parkinson’s disease based on the detection of dysphonia. New methods are presented for enhanced evaluation and recognize Parkinson’s disease affected patients at early stage. Analysis is performed with significant level of error tolerance rate and established our results with corrected T-test. Here new ensembles and other machine learning methods consisting of mult….