The graphical feature extraction of star plot for wine quality classification

摘要

we propose a visualization method of evaluation of wine quality. The wine data are from the certification phase of the physicochemical analysis test. The data include the 11 input variables, an output variable which is the quality of wine. The data include 1599 samples of red wine and 4898 samples of white wine. Our method works better than the traditional neural networks and support vector machine method, and has visual advantages. Such model is useful to support the oenologist wine tasting evaluations and improve wine production. Furthermore, similar techniques can help in target marketing by modeling consumer tastes from niche markets.