Using orange data mining for meat classification: The preliminary application of machine learning
Issued Date
2024-11-01
Resource Type
eISSN
26300192
Scopus ID
2-s2.0-85216287437
Journal Title
International Journal of Agricultural Technology
Volume
20
Issue
6
Start Page
2497
End Page
2512
Rights Holder(s)
SCOPUS
Bibliographic Citation
International Journal of Agricultural Technology Vol.20 No.6 (2024) , 2497-2512
Suggested Citation
Phoemchalard C., Senarath N., Malila P., Tathong T., Khamhan S. Using orange data mining for meat classification: The preliminary application of machine learning. International Journal of Agricultural Technology Vol.20 No.6 (2024) , 2497-2512. 2512. Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/104242
Title
Using orange data mining for meat classification: The preliminary application of machine learning
Author(s)
Author's Affiliation
Corresponding Author(s)
Other Contributor(s)
Abstract
Orange Data Mining study on the classification of buffalo, beef, and goat meats, Machine Learning (ML) classifiers including Support Vector Machine (SVM), Neural Network (NN), and Naïve Bayes (NB) are well performed to achieve 100% accuracy across all features. Random Forest (RF) demonstrated the best performance more than 97% in AUC, CA, F1, and MCC. Other models such as Gradient Boosting (GB), AdaBoost, CN2 Rule Induction (CN2), Decision Tree (DT), and k-Nearest Neighbors (KNN) are performed better but there were less efficient. In the application of specific classifiers for species-based meat quality attributes, SVM, NN and NB should be considered as the best options.