Machine Learning based Wine Quality Evaluation

Abstract

In the wine industry, red wine is a kind of fruit wine that made from grapes, which is quite common in our life. With the gradual expansion of people who like red wine, the wine industry is getting more and more attention at the same time, the quality of red wine is also getting more and more attention. In order to better evaluate the quality of red wine, an evaluation model of red wine quality is established based on the collected 200 red wine samples and the corresponding 11 index data. Firstly, the dimensionality of the data index is decreased by using principal component analysis and factor analysis. Then, k-means clustering, sum of squares of deviation and class average algorithm are used to perform cluster analysis on the data processed by principal component analysis or factor analysis. Then, logistic regression analysis is used to test the accuracy of the data classification. Finally, Fisher discriminant method is used to perform discriminant analysis on the data and establish a model. The score function is obtained, and the score function is used to calculate the quality score of each group of wine, and the corresponding suggestions were given according to the quality evaluation results.