Data Mining im Verkehrswesen.

Abstract

Data Mining kann einen Beitrag zur Wissensgenerierung aus Daten liefern. Es wird ein Überblick über Methoden und Verfahren des Data Minings gegeben. Anhand von Beispielen wird gezeigt, dass diese Methoden und Verfahren auch im Verkehrswesen in vielen Fällen eingesetzt werden können. difu'Data mining' methods permit the examination of extensive data bases to extract knowledge, thus offering analytical support for statistical evaluations. This is of special interest in those instances where links between the available data - and the corresponding models of evaluation - are not sufficiently known. In addition to a brief introduction to 'data mining', its application in the field of transport is explained with the help of examples. 'Data mining' offers methods of forecasting (classification / regression), grouping together of selected data (clustering), and the uncovering of interdependent issues and rules. This may be related to actual knowledge / experience or data bases, and comprise 'Fuzzy Technology', 'Neuronal Networks' and 'Decision Tree structures'. 'Fuzzy Technology' is based on expert knowledge while 'Multilayer Perceptron', 'Fuzzy C-Means' (both neuronal networks), and 'Decision Tree' methods are linked to data bases.