Fuzzy Information and Data Analysis

Abstract

The results of data warehousing and data mining are depending essentially on the quality of data. Usually data are assumed to be numbers or vectors, but this is often not realistic. Especially the result of a measurement of a continuous quantity is always not a precise number, but more or less non-precise. This kind of uncertainty is also called fuzziness and should not be confused with errors. Data mining techniques have to take care of fuzziness in order to avoid unrealistic results.