Boosted Stacking Ensemble Machine Learning Method for Wafer Map Pattern Classification

Recently, machine learning-based technologies have been developed to automate the classification of wafer map defect patterns during semiconductor manufacturing. The existing approaches used in pattern include directly learning image through a convolution neural network and applying ensemble method after extracting features. This study aims classify defects more effectively derive robust algorithms even for datasets with insufficient patterns. First, number actual process may be limited. Therefore, data are generated using convolutional auto-encoder (CAE), expanded verified evaluation technique structural similarity index measure (SSIM). After handcrafted features, boosted stacking model that integrates four base-level classifiers extreme gradient boosting classifier as meta-level is designed built training based on final prediction. Since proposed algorithm shows better performance than those patterns, results this will contribute improving product quality yield manufacturing process.