This paper presents a statistical approach for rule-base generation of
handwriting recognition. The proposed method integrates the heuristic
feature selection with the statistical evaluation and thus improves t
he performance of the rule generation as well as of the fuzzy handwrit
ing recognition system. Fuzzy statistical measures are employed to ide
ntify relevant features from a given large handwriting database. First
an automatic rule-base mechanism is presented. The reduce the time ne
eded for this generation mechanism an additional heuristic feature sel
ection step is introduced. Tests show that this generated rule-base im
proved the recognition results over previous approaches.