A Machine Learning Based Approach for Wine Quality Prediction
Basvaraj. S. Anami, Kavita Mainalli, Shanta Kallur, Vijeeta Patil
Abstract
Wine is a popular drink across the globe and the gender. Older the Wine, better is the taste but, expensive. The Wine quality is measured based on the important parameters, such as free Sulphur dioxide, Volatile acidity, Citric Acid and Residual sugar. The traditional way of Wine quality assessment was time consuming. This paper gives an automatic prediction of Wine quality, as good or bad, using machine learning approaches which are Neural Networks, Logistic Regression and Support Vector Machine are implemented on standard datasets of Portuguese "Vinho Verde" Wine. The results are compared with standard values. The support vector Machine has achieved superior than other techniques with error of 0.003. The quality rate for SVM is 7.99. The work is useful in Wine industry for quality testing and assurance for customers.