Wine Quality Prediction using CNN

TITLE	Wine Quality Prediction using CNN
ABSTRACT	Wine quality assessment is essential to ensure customer satisfaction and maintain production standards. Traditional methods by tasters are subjective and inconsistent. To address this, machine learning (ML) provides an automated and reliable approach using physicochemical properties like acidity, pH, alcohol, sulphates, and residual sugar. This project develops a predictive model to classify wine quality as good or bad using supervised algorithms. Data preprocessing techniques such as handling missing values, normalization, and feature selection are applied to enhance performance. Models like Decision Tree, Random Forest, SVM, and KNN are compared using accuracy, precision, recall, and F1-score. Results show that Random Forest outperforms single models, offering higher accuracy. The proposed system helps winemakers in quick, consistent, and efficient quality evaluation.
AUTHOR	Jayanth Mane, Adarsh M J
PUBLICATION DATE	2025-08-28 21:00:05
VOLUME	13
ISSUE	3
DOI	DOI: 10.15662/IJMSERH.2025.1303058
PDF	pdf/2025/7/58_Wine Quality Prediction using CNN.pdf
KEYWORDS

Article