A Fuzzy System for Combining Different Outliers Detection Methods

S. Cateni, V. Colla, and M. Vannucci (Italy)


Outliers detection, fuzzy inference system


The paper presents an application of fuzzy inference system for outliers detection. The basic idea of the proposed method lies in the combination of different traditional outliers detection methods in order to exploit all their advantages by overcoming their possible limitations. The developed method requires neither a priori assumptions on the processed data nor preliminary statistical analyses or parameters tuning and setup made by the users according to possibly subjective criteria. The proposed method have been tested in an industrial context, in order to filter out unreliable measurements in data series used for process control. The results obtained in the processing of experimental data are presented and discussed.

Important Links:

Go Back