Vol. 2 No. 1 (2023): JOURNAL OF SCIENCE, ENGINEERING AND TECHNOLOGY YABATECH
Articles

AN EFFECTIVE MACHINE LEARNING MODEL FOR WINE CLASSIFICATION

PDF
  • Okikiola, F. M.,
  • Akinola, A. F,
  • Ishola, P. E.,
  • Onadokun, I .O.
Okikiola, F. M.
Department of Computer Technology,Yaba College of Technology, Yaba, Lagos State, Nigeria
Bio
Akinola, A. F
Department of Computer Technology,Yaba College of Technology, Yaba, Lagos State, Nigeria
Bio
Ishola, P. E.
Department of Computer Technology,Yaba College of Technology, Yaba, Lagos State, Nigeria
Bio
Onadokun, I .O.
Department of Computer Technology,Yaba College of Technology, Yaba, Lagos State, Nigeria
Bio

Published 2023-10-19

Keywords

  • Decision trees, Random forests, stochastic gradient descent and Wine.

How to Cite

Okikiola, F. M., Akinola, A. F, Ishola, P. E., & Onadokun, I .O. (2023). AN EFFECTIVE MACHINE LEARNING MODEL FOR WINE CLASSIFICATION. Journal of Science, Engineering and Technology YABATECH, 2(1). Retrieved from https://josetyabatech.com/index.php/home/article/view/20

Abstract

Wine companies spend a lot of money to test the quality of their wine because they need to buy

specialized equipment and build elaborate winery labs to house it. Lab testing also takes a lot of

time because it is so labour-intensive. Some people even go so far as to hire qualified taste

consultants, which is an expensive alternative. In this instance, we created an effective machine

learning model that can forecast wine quality based on some physicochemical traits, enabling the

production and distribution of the highest quality wine. Using decision trees, random forests, and

stochastic gradient descent. This was accomplished by using an industry-specific database for the

wine-making industry. It was shown that the random forest strategy outperformed the other two

strategies after training and testing on a set of dataa higher accuracy of 93%. This shows how

wine companies can start saving money and making decisions that are much more informed.

PDF