Convolutional Neural Network for Halal Detection of Korean Cosmetic Composition

Korean cosmetics occupy the position as the best and most favorite cosmetics in Indonesia with a user percentage of 46.6%, beating domestic cosmetics with 34.1%. Unfortunately, Hangeul's writing on Korean cosmetic packaging often confuses the contents of the cosmetics. In fact, as a country with the most significant Muslim majority in the world, Indonesian people are required to use everything halal. A halal detection application for Korean cosmetic compositions was created by implementing the Convolutional Neural Network. The test results show that the application can detect material doubts with an accuracy rate of 95.56%. This indicates that the Korean cosmetic halal detection application is in a suitable category.