Sample Efficient Regression Trees (SERT) for Yield Loss Analysis

Abstract

Regression trees have been known to be an effective data mining tool for semiconductor yield analysis. The regression tree is built by iteratively splitting data set and selecting factors into a hierarchical tree model. The sample size reduces sharply after few levels of data splitting and causes unreliable factor selection. In contrast, the forward regression analysis selects the influential variables all the way with the same set of data. Regression analysis is, however, not capable of splitting data into groups with different models. In this research, we propose a sample- efficient regression tree (SERT) that combines the forward regression and regression tree methodologies and show that SERT is effective in discovering yield-loss causes during the yield ramp-up stage where the sample size available for analysis is extremely small.