Implementasi Support Vector Machine dan Radial Basis Function untuk Klasifikasi Makanan Vegetarian Menggunakan Data Image

Williams (Universitas Universal)
Fery Gunawan (Universitas Universal)
Patrick Limuel (Universitas Universal)
Akhmad Rezki Purnajaya (Universitas Universal)

JoDENS: Journal of Digital Ecosystem for Natural Sustainability
Vol 3 No 1 (2023): Juli 2023

The vegetarian diet has become increasingly popular in the 21st century due to its potential to reduce the risk of chronic and degenerative diseases. Vegetarians are individuals who do not consume animal products, either for religious or health reasons. However, it can be difficult to determine whether a particular food is vegetarian or non-vegetarian based on visual inspection alone. Therefore, this study successfully developed an SVM & RBF model in RStudio that can accurately differentiate between vegetarian and non-vegetarian foods based on image data. The model achieved an accuracy rate of 95%, specificity of 100%, sensitivity of 88.89%, and an AUC value of 94.44%. It can be concluded that the SVM & RBF model is capable of predicting data with high accuracy and effectively distinguishing between vegetarian and non-vegetarian classes.