Extracting knowledge and computable models from data - Needs, expectations, and experience
|T. Natschläger, F. Kossak, M. Drobics. Extracting knowledge and computable models from data - Needs, expectations, and experience. pages 493 - 498, 7, 2004.|
|Buch||Proc. 13th IEEE Int. Conf. on Fuzzy Systems|
|Seiten||493 - 498|
In modern industrial manufacturing, a great amount of data is gatheredto monitor and analyze a given production process. Intelligentanalysis of such data helps to reveal as much information about theproduction process as possible. This information is most useful if itis available in the form of interpretable and predictive models. Suchmodels can be generated from data by means of (fuzzy logic based)machine learning methods. In this contribution we will describeindustrial applications in the areas of process optimization andquality control where we have successfully establishedmachine-learning methods as intelligent data analysis tools. Based onthese applications, we will report the characteristics ofmachine-learning tools which according to our experience supportsuccessful applications in an industrial environment. Furthermore wedescribe some methodical aspects resulting from this applications.