Facial Skin Type Prediction Based on Baumann Skin Type Solutions Theory Using Machine LearningFacial
Facial Skin Type Prediction Based on Baumann Skin Type Solutions Theory Using Machine Learning
Author(s): Rosayanti Efata, Widya Indriani Loka, Natasha Wijaya, Derwin SuhartonoSubject(s): Education and training, ICT Information and Communications Technologies
Published by: UIKTEN - Association for Information Communication Technology Education and Science
Keywords: Skin type classification; Baumann Skin Type Solutions; XGBoost, SVM; 1D-CNN; Bayesian Optimization
Summary/Abstract: The lack of knowledge of different facial skin types is still a frequent problem in Indonesia. The purpose of this research is to build a facial skin type prediction system using machine learning to classify facial skin types based on Baumann Skin Type Solutionswhich provides information on different skin types and suitable skincare ingredients. The dataset is collected manually by distributing a questionnaire among Indonesian citizens. The prediction models are built using three machine learning methods namely SVM, XGBoost, and 1D-CNN, and compared using 5-fold stratified cross-validation. XGBoostachieved the best performance on facial skin type prediction and optimized through hyperparameter tuning using Bayesian Optimization with a result of 93.5% averaged F1-score.
Journal: TEM Journal
- Issue Year: 12/2023
- Issue No: 1
- Page Range: 96-103
- Page Count: 8
- Language: English