An automatic rule base generation method for fuzzy pattern recognition with multiphased clustering

Presents an approach for the automatic generation of fuzzy rule bases for pattern recognition from a given sample data. The general idea of the approach is to use and enhance the fuzzy c-means clustering algorithm. The rule base is generated through a modified iterative feature clustering method. A following cross-checking is used to separate the generated rules. Although the rule base generation method was initially developed for handwriting features the scope of its applicability is much larger. The proposed clustering algorithm was tested with input feature space up to 125 dimensions.