Nowadays, product quality certificates are being used by enterprises to market their products. This is an expensive and ineffective process that takes a long time and requires assessments from customers and human experts. Determining the relationship between a red wine's chemical composition and subjective quality is a difficult task. This paper investigates the use of statistical methods, such as linear regression, to predict the values of target variables and ascertain how dependent the target variable is on the independent factors. It is more accurate than previous methods. These results contribute to the understanding of how different red wine consumers perceive quality and can help the red wine industry identify the primary sensory-active ingredients that influence quality in different red wines. The paper uses linear regression to solve the problem. The research examines at those factors' VIF value and significance in order to assess the efficacy of this operation. It turns out that volatile acidity, chlorides, total sulfur dioxide, sulphates and alcohol have a significant linear relationship with red wine quality, while the other factors have less influence on red wine quality. Overall, red wine quality can be evaluated based on how much these variables affect them.
Read full abstract