이상치를 고려한 딥러닝 기반 웨이퍼 맵 결함 패턴 클러스터링

Abstract

Purpose: Most clustering algorithms perform clustering without considering outliers. However, when performing clustering, outliers often degrade performance. Therefore, we propose a clustering method that can consider outliers.BRMethods: The proposed deep learning-based algorithm identifies outliers using the predicted class distribution and performs clustering based on only normal data.BRResults: The proposed algorithm achieved superior performance compared to the existing clustering algorithm. Based on the proposed clustering algorithm, the wafer maps were clustered while finding outliers in the patterns existing on them.BRConclusion: In this paper, we propose a deep learning-based clustering algorithm considering outliers (DCCO) that simultaneously performs clustering and abnormal pattern detection.