본문 바로가기
KERD
논문KCI 등재

LIME을 활용한 준지도 학습 기반 이상 탐지 모델: 반도체 공정을 중심으로

Anomaly Detection Model Based on Semi-Supervised Learning Using LIME: Focusing on Semiconductor Process

신주은|백동현|안강민
한국산업경영시스템학회 (KOREAN SOCIETY OF INDUSTRIAL AND SYSTEMS ENGINEERING)|2022.12|한국산업경영시스템학회지 (Journal of Korean Society of Industrial and Systems Engineering)|제45권 제4호|pp. 86-98

0

조회

0

다운로드

0

인용

0

참고문헌

키워드

LIMESMOTEAnomaly DetectionSemiconductor Fabrication ProcessRFECV
주제분류산업공학

초록

Recently, many studies have been conducted to improve quality by applying machine learning models to semiconductor manu- facturing process data. However, in the semiconductor manufacturing process, the ratio of good products is much higher than that of defective products, so the problem of data imbalance is serious in terms of machine learning. In addition, since the number of features of data used in machine learning is very large, it is very important to perform machine learning by extracting only important features from among them to increase accuracy and utilization. This study proposes an anomaly detection method- ology that can learn excellently despite data imbalance and high-dimensional characteristics of semiconductor process data. The anomaly detection methodology applies the LIME algorithm after applying the SMOTE method and the RFECV method. The proposed methodology analyzes the classification result of the anomaly classification model, detects the cause of the anomaly, and derives a semiconductor process requiring action. The proposed methodology confirmed applicability and feasibility through application of cases.

Abstract

Recently, many studies have been conducted to improve quality by applying machine learning models to semiconductor manu- facturing process data. However, in the semiconductor manufacturing process, the ratio of good products is much higher than that of defective products, so the problem of data imbalance is serious in terms of machine learning. In addition, since the number of features of data used in machine learning is very large, it is very important to perform machine learning by extracting only important features from among them to increase accuracy and utilization. This study proposes an anomaly detection method- ology that can learn excellently despite data imbalance and high-dimensional characteristics of semiconductor process data. The anomaly detection methodology applies the LIME algorithm after applying the SMOTE method and the RFECV method. The proposed methodology analyzes the classification result of the anomaly classification model, detects the cause of the anomaly, and derives a semiconductor process requiring action. The proposed methodology confirmed applicability and feasibility through application of cases.

참고문헌 (49)

  1. Al Sarah, N, "An Efficient Android Malware Prediction Using Ensemble machine learning algorithms", Procedia Computer Science, 191, pp.184~191 (2021)
  2. An, J. H, "XAI, Explanable Artificial Intelligence, Dissects Artificial Intelligence" (2020), Wikibooks
  3. "Developing and Evaluating an Anomaly Detection System"[URL]
  4. Breiman, L, "Random Forests", Machine Learning, 45, pp.5~32 (2001)
  5. Chandola, V, "Anomaly detection : A survey", ACM computing surveys(CSUR), 41(3), pp.1~58 (2009)
  6. Chauhan, K. K, "Semiconductor wafer defect classification using convolution neural network: a binary case", IOP Conference Series: Materials Science and Engineering, 1225(1), pp.012060~ (2022), IOP Publishing
  7. Chawla, N. V, "SMOTE : synthetic minority over-sampling technique", Journal of Artificial Intelligence Research, 16, pp.321~357 (2002)
  8. Choi, S. J, "Technology of Design and Manufacturing Process of Nano Semiconductor Devices" (2021), Free Academy
  9. Cox, D.R, "The regression analysis of binary sequences", Journal of the Royal Statistical Society: Series B (Methodological), 20(2), pp.215~232 (1958)
  10. Doran, D, "What does explainable AI really mean? A new conceptualization of perspectives", arXiv preprint arXiv:1710.00794, ~ (2017)
  11. Ertekin, S, "Learning on the border: active learning in imbalanced data classification", Proceedings of the Sixteenth ACM Conference on Conference on Information and Knowledge Management, pp.127~136 (2007)
  12. Goodlin, B. E, "Simultaneous fault detection and classification for semiconductor manufacturing tools", Journal of the Electrochemical Society, 150(12), pp.G778~ (2003)
  13. Guyon, I, "Gene selection for cancer classification using support vector machines", Machine Learning, 46(1), pp.389~422 (2002)
  14. Hawkins, D.M, "Identification of outliers", Biometrical Journal, 29, pp.198~198 (1980), Chapman and Hall
  15. 허선우, "데이터 불균형을 고려한 설명 가능한 인공지능 기반 기업부도예측 방법론 연구", 한국산업경영시스템학회지, 45(2), pp.65~76 (2022), 한국산업경영시스템학회
  16. Jung, I.S, "The Future of the Semiconductor Empire" (2021), Ire media
  17. Ke, G, "Lightgbm : A Highly Efficient Gradient Boosting Decision Tree", Advances in Neural Information Processing Systems, 30, pp.3146~3154 (2017)
  18. Kim, C.G, "LSTM based Anomaly Detection on semiconductor manufacturing data", Proceedings of the Korean Information Science Society Conference, pp.760~762 (2017)
  19. 김동욱, "이상 탐지 분석에서 알려지지 않는 공격을 식별하기 위한 이산 웨이블릿 변환 적용 연구", 인터넷정보학회논문지, 22(3), pp.45~52 (2021), 한국인터넷정보학회
  20. 김희수, "생성적 적대 신경망을 이용한 반도체 제조공정 데이터의 결측치 추정 및 공정 이상 진단 프레임웍", 한국지능시스템학회 논문지, 28(4), pp.393~400 (2018), 한국지능시스템학회
  21. 김종은, "수질자료의 이상치 탐색을 위한 Isolation Forest기법의 적용", 대한환경공학회지, 40(12), pp.473~480 (2018), 대한환경공학회[DOI]
  22. Kim, J. K, "Data imbalance problem solving for smote based oversampling: Study on fault detection prediction model in semiconductor manufacturing process", Advanced Science and Technology Letters, 133, pp.79~84 (2016)
  23. 김정우, "반도체 설비의 효율성 제고를 위한 설비 할당 스케줄링 규칙에 관한 연구", 한국산업경영시스템학회지, 39(1), pp.153~161 (2016), 한국산업경영시스템학회
  24. Kim, J. W, "Strategies to leverage manufacturing big data" (2020), haum
  25. "Silicon Times, Vol. 601" (2021)[URL]
  26. Kwon, C.M, "Python Machine Learning Complete Guide" (2020), Wikibooks
  27. 이장희, "반도체 제조공정에서의 이상수율 검출 방법론", Journal of Information Technology Applications & Management, 15(1), pp.243~260 (2008), 한국데이터전략학회
  28. 이용준, "로지스틱 회귀분석 및 AHP 기법을 이용한 산사태 위험지역 분석", 대한토목학회 논문집D, 26(5D), pp.861~867 (2006), 대한토목학회
  29. Li, J, "Feature selection: A data perspective", ACM computing surveys (CSUR), 50(6), pp.1~45 (2017)
  30. Liao, D.Y, "Anomaly detection for semiconductor tools using stacked autoencoder learning", 2018 International Symposium on Semiconductor Manufacturing (ISSM), pp.1~4 (2018), IEEE
  31. Liu, F.T, "Isolation Forest", 2008 Eighth IEEE International Conference on Data Mining, pp.413~422 (2008)
  32. Liu, J, "A comparative study on rough set based class imbalance learning", Knowledge-Based Systems, 21(8), pp.753~763 (2008)
  33. Maggipinto, M, "A Deep Convolutional Autoencoder-Based Approach for Anomaly Detection With Industrial, Non-Images, 2-Dimensional Data: A Semiconductor Manufacturing Case Study", IEEE Transactions on Automation Science and Engineering (2022)
  34. McKinsey & Company, "Game changers: Five opportunities for US growth and renewal" (2013), McKinsey Global Institute
  35. "UCI Machine Learning Repository" (2008)[URL]
  36. 남충현, "LIME 알고리즘을 이용한 한국어 감성 분류 모델 해석", 한국정보통신학회논문지, 25(12), pp.1784~1789 (2021), 한국정보통신학회[DOI]
  37. Randolph-Gips, M, "A new neural network to process missing data without Imputation", 2008 Seventh International Conference on Machine Learning and Applications, pp.756~762 (2008), IEEE
  38. Raschka, S, "Machine Learning Textbook with Python, Scikit-Learn, TensorFlow", pp.137~140 (2019), gilbut
  39. Ribeiro, M. T, "Why should i trust you?” Explaining the predictions of any classifier", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.1135~1144 (2016)
  40. Rothman, D, "Hands-On Explainable AI(XAI) with Python" (2021), DK Road Books
  41. Samek, W, "Explainable artificial intelligence: Understanding, visualizing and interpreting deep learning models", arXiv preprint arXiv:1708.08296, ~ (2017)
  42. Schlosser, T, "Improving automated visual fault inspection for semiconductor manufacturing using a hybrid multistage system of deep neural networks", Journal of Intelligent Manufacturing, pp.1~25 (2022)
  43. Schölkopf, B, "Estimating the support of a high-dimensional distribution", Neural Computation, 13(7), pp.1443~1471 (2001)
  44. "Explainable AI" (2021)[URL]
  45. Stehman, S.V, "Selecting and interpreting measures of thematic classification accuracy", Remote sensing of Environment, 62(1), pp.77~89 (1997)
  46. Stekhoven, D. J, "MissForest—non-parametric missing value imputation for mixed-type data", Bioinformatics, 28(1), pp.112~118 (2012)
  47. Susto, G. A, "Anomaly detection approaches for semiconductor manufacturing", Procedia Manufacturing, 11, pp.2018~2024 (2017)
  48. West, D, "Neural network ensemble strategies for financial decision applications", Computers and Operations Research, 32(10), pp.2543~2559 (2005)
  49. "XGBoost Tutorials - xgboost 1.4.0-SNAPSHOT documentation"[URL]

관련 논문AI추천

반도체 제조공정 품질 향상을 위한 그래프 신경망 기반 이상탐지 및 근본 원인 시각화 사례 연구김두리, 이승준, 최정일, 하주은한국품질경영학회 · 품질경영학회지2025 · 제53권 제2호SVM 기반 Bagging과 OoD 탐색을 활용한제조공정의 불균형 Dataset에 대한 예측모델의 성능향상김종훈, 오하영한국정보처리학회 · 정보처리학회 논문지2022 · 제11권 제11호머신러닝을 이용한 반도체 웨이퍼 평탄화 공정품질 예측 및 해석 모형 개발안정언, 정재윤사)한국빅데이터학회 · 한국빅데이터학회 학회지2019 · 제4권 제2호CNN을 활용한 웨이퍼 불량 원인 인자 파악에 관한 연구 : 반도체 전공정 중심으로김학수, 성대식한국상업경영학회 · 상업경영연구2022 · 제36권 제2호불량품 예측을 위한 이상치 탐지 기법 연구김태국, 이태오한국사물인터넷학회 · 사물인터넷융복합논문지2025 · 제11권 제6호제조 공정 결함 탐지를 위한 MixMatch 기반 준지도학습 성능 분석김예준, 김용수, 정예은한국산업경영시스템학회 · 한국산업경영시스템학회지2023 · 제46권 제4호

이 간행물의 다른 논문 (27건)

텍스트 기반 Explainable AI를 적용한 국가연구개발혁신 모니터링배성훈, 임정선pp. 1-7전투기 도장 제거용 로봇형 레이저 코팅 제거 시스템 도입 의사결정을 위한 증분분석이창용, 박종훈pp. 8-20블랙 숄즈 모델을 이용한 잠재적 풍력발전 위치의 경제적 가치분석심재훈pp. 21-30게스케(Geske) 모델을 이용한 신재생에너지사업의 경제성 분석심재훈pp. 31-41다종 장사정포 공격에 대한 강화학습 기반의 동적 무기할당김정훈, 경지훈, 김현호, 공주회pp. 42-52
원문 보기