Research on Red Wine Quality Based on Data Visualization

Due to the rapid development of modern society, red wine has gradually become popular. Research on the quality of red wine has turn into an important topic. Red wine contains more than 600 kinds of ingredients, in terms of alcohol, minerals, tannic acid, citric acid, chloride and other substances. This paper analyzes 12 factors affecting red wine quality in the data set and studies the influence of each ingredient on the quality of red wine through data mining algorithm. Based on the data visualization of Python processing, classical visualization tools such as histogram, heat map, box-plot and Pearson correlation coefficient algorithm are used for data mining. The histogram is adopted for univariate analysis and the heat map composed of Pearson coefficient is used for multivariate analysis. Then, the box-plot is used for cross-verification. Finally, it is concluded that alcohol, sulfate, citric acid and volatile acidity are the decisive factors affecting the quality of red wine. This conclusion can not only be used as a reference for consumers to buy, but also provide suggestions for wine manufacturers to improve the quality of red wine.