Designing A Wine Quality Identification Application Using Naïve Bayes Classifier

Wine is one of the drinks with the highest number of consumers in the world, and its quality is greatly influenced by various chemical ingredients such as fixed acidity, citric acid, pH, and others. Wine quality testing is traditionally subjective and requires special expertise. Therefore, a more objective and efficient approach is needed, one of which is through the application of data mining algorithms. This study aims to design a wine quality identification application using the Naïve Bayes Classifier algorithm and analyze its accuracy in the classification of red and white wines. The dataset used was taken from the UCI Machine Learning Repository, with input attributes in the form of wine chemical content and output in the form of wine quality scores ranging from 2 to 8. The test results showed that the Naïve Bayes algorithm was able to classify the quality of wine with an accuracy rate of 86.50% for red wines and 79.80% for white wines. The developed application successfully processes input data and provides prediction results in real-time. This research shows that the use of Naïve Bayes algorithms can be a practical and effective solution in helping consumers and producers in recognizing the quality of wine and improving the production process in the wine industry.