Title: Classifications on wine informatics using PCA, LDA, and supervised machine learning techniques

Authors: Swarna Prabha Jena; Bijay Kumar Paikaray; Jitendra Pramanik; Rishi Thapa; Abhaya Kumar Samal

Addresses: Department of ECE, Centurion University of Technology and Management, Odisha, India ' School of Information and Communication Technology, Medhavi Skills University, Sikkim, India ' Department of Mining, National Institute of Technology, Rourkela Odisha, India ' School of Management, Amity University, Madhya Pradesh, India ' Department of CSE, Trident Academy of Technology, Bhubaneswar, Odisha, India

Abstract: Proving the quality of a food product is challenging for any country. Every country recommends using products whose quality has been assured. A similar thing applies to the wine industry. To promote their products, wine industries acquire quality certifications through expert assessments. It is an expensive and time-consuming process. This paper explores the usage of machine algorithms like principle component analysis (PCA), linear discriminant analysis (LDA), random forest (RF), Gaussian naive Bayes (GNB), decision trees (DT), K-nearest neighbour (KNN), logistic regression (LR), and gradient boost (GB) for classifying the wine data into three main categories. The experimental work provides a comparative study of the accuracy of all classifiers is discussed in detail.

Keywords: classification; machine learning; decision tree; principle component analysis; PCA; linear discriminant analysis; LDA; Gaussian naive Bayes; GNB; K-nearest neighbour; KNN.

DOI: 10.1504/IJWI.2023.130444

International Journal of Work Innovation, 2023 Vol.4 No.1, pp.58 - 73

Received: 04 Jan 2023
Accepted: 30 Jan 2023
Published online: 20 Apr 2023 *