Enhancing Confidence and Interpretability of a CNN-Based Wafer Defect Classification Model Using Temperature Scaling and LIME
Micromachines · 17.09.2025 · 16 (9) · 1057
DOI: 10.3390/mi16091057 ↗Kurz gefasst
Fachartikel in der Zeitschrift Micromachines, online erschienen am 17.09.2025. Verfasst von Jieun Lee und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of System Semiconductor Engineering, Sangmyung University, Cheonan 31066, Republic of Korea. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Sangmyung University.
Abstract (Originalsprache)
Accurate classification of defects in the semiconductor manufacturing process is critical for improving yield and ensuring quality. While previous works have mainly focused on improving classification accuracy, we propose a model that can simultaneously assess accuracy, prediction confidence, and interpretability in wafer defect classification. To solve the class imbalance problem, we used a weighted cross-entropy loss function and convolutional neural network–based model to achieve a high accuracy of 97.8% on the test dataset and applied a temperature-scaling technique to enhance confidence. Furthermore, by simultaneously employing local interpretable model-agnostic explanations and gradient-weighted class activation mapping, the rationale for the predictions of the model was visualized, allowing users to understand the decision-making process of the model from various perspectives. This research can provide a direction for the next generation of intelligent quality management systems by enhancing the applicability of the proposed model in actual semiconductor production sites through explainable predictions.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of System Semiconductor Engineering, Sangmyung University, Cheonan 31066, Republic of Korea.
- Department of Human Intelligence and Robot Engineering, Sangmyung University, Cheonan 31066, Republic of Korea.
ORCID-Profile: Soo-Whang Baek, Jonghwan Lee
Förderung
- Sangmyung University
Zitieren
APA
Lee, J., Ju, Y., Lim, J., Hong, S., Baek, S. W., & Lee, J. (2025). Enhancing Confidence and Interpretability of a CNN-Based Wafer Defect Classification Model Using Temperature Scaling and LIME. Micromachines, 16(9), 1057. https://doi.org/10.3390/mi16091057
Vancouver
Lee J, Ju Y, Lim J, Hong S, Baek SW, Lee J. Enhancing Confidence and Interpretability of a CNN-Based Wafer Defect Classification Model Using Temperature Scaling and LIME. Micromachines. 2025;16(9):1057. doi:10.3390/mi16091057
BibTeX
@article{Lee2025,
author = {Lee, Jieun and Ju, Yeonwoo and Lim, Junho and Hong, Sungmin and Baek, Soo-Whang and Lee, Jonghwan},
title = {Enhancing Confidence and Interpretability of a CNN-Based Wafer Defect Classification Model Using Temperature Scaling and LIME},
journal = {Micromachines},
year = {2025},
volume = {16},
number = {9},
pages = {1057},
doi = {10.3390/mi16091057}
}
EUVIMED