Identifying Key Features for Quality Prediction from Production Data
Chung-Chian Hsu, Yi-Ling Xiao, Guanlin Chen, Arthur Chang, An-Yi Hsu · 2024
This study aims to use machine learning to predict the quality of hot stamping foil products and to identify the subset of input features that has a significant impact on the prediction. With the development of artificial intelligence, domain experts are actively seeking effective methods to monitor production process. This study proposes a framework which uses state-of-the-art feature selection technique to identify the best feature subset and hyperparameter tuning for training advanced machine learning models to predict product quality. Experimental results on a real-world dataset demonstrate feasibility of the proposed framework.