Wine Quality Prediction using Machine Learning

This paper attempts to explore the prediction of wine quality by several machine learning algorithms such as K- Nearest Neighbours (KNN), Random Forest (RF), Logistic Regression (LR), Support Vector Classification (SVC), and Decision Trees classifier (DTC). These models achieved different accuracy scores. The Random Forest model was highly effective and achieved the highest score, at 91.5%. The KNN model, although lower at 88.5%, was powerful in its prediction power. The SVC and LR models were also very impressive at 90.8% and 89.8%, respectively. Therefore, the bottom line of this study is that feature selection and the choice of a model matter in increasing the chances of an accurate prediction for assessing wine quality.