A clustering assisted method for fuzzy rule extraction and pattern classification
H. Kuo, T.D. Gedeon, Patrick M. Wong · 2003
The fuzzy classification method developed by S. Abe and M.S. Lan (1995) has been improved. This method extracts fuzzy rules directly from numerical data. The paper shows how preprocessing input data using clustering may help the classification accuracy in some cases. The proposed method is compared with Abe and Lan's fuzzy classification method with a data set obtained from an oil reservoir in the North West Shelf in Australia and the Fisher Iris data.