Application of f-regression method to fuzzy classification problem

In regression analysis, outliers always represent difficulties because they cause modeling errors. But under certain circumstances, they can actually contain useful information, as shown on the example of problem described in this article. That is why the task of outlier identification and analysis presents a twofold interest from the technical point of view. It is shown that f-regression method has good outlier detection capability and can be successfully applied to fuzzy classification problem.

• Type: Conference
• Year: 2003
• Where: EUSFLAT
• Authors: Boris Izyumov
• Keywords: EUSFLAT 2003, EUSFLAT 2007, Outlier, Outlier Detection Capability, Outlier Identification