<> "The repository administrator has not yet configured an RDF license."^^ . <> . . "Data Engineering for the Analysis of Semiconductor Manufacturing Data"^^ . "We have analyzed manufacturing data from several different semiconductor \nmanufacturing plants, using decision tree induction software called \nQ-YIELD. The software generates rules for predicting when a given product \nshould be rejected. The rules are intended to help the process engineers \nimprove the yield of the product, by helping them to discover the causes \nof rejection. Experience with Q-YIELD has taught us the importance of \ndata engineering -- preprocessing the data to enable or facilitate \ndecision tree induction. This paper discusses some of the data engineering\nproblems we have encountered with semiconductor manufacturing data.\nThe paper deals with two broad classes of problems: engineering the features \nin a feature vector representation and engineering the definition of the \ntarget concept (the classes). Manufacturing process data present special \nproblems for feature engineering, since the data have multiple levels of \ngranularity (detail, resolution). Engineering the target concept is important, \ndue to our focus on understanding the past, as opposed to the more common \nfocus in machine learning on predicting the future."^^ . "1995" . . . . . . . . "Peter"^^ . "Turney"^^ . "Peter Turney"^^ . . . . . . "Data Engineering for the Analysis of Semiconductor Manufacturing Data (PDF)"^^ . . . . . . "NRC-39163.pdf"^^ . . . "Data Engineering for the Analysis of Semiconductor Manufacturing Data (Image (PNG))"^^ . . . . . . "preview.png"^^ . . . "Data Engineering for the Analysis of Semiconductor Manufacturing Data (Indexer Terms)"^^ . . . . . . "indexcodes.txt"^^ . . "HTML Summary of #2891 \n\nData Engineering for the Analysis of Semiconductor Manufacturing Data\n\n" . "text/html" . . . "Machine Learning" . . . "Artificial Intelligence" . .