A new recognition system based on a neuro-fuzzy system, called FasArt, is p
roposed in this paper. Satisfactory results were obtained using the train_r
01_v02 UNIPEN dataset, together with a comparison with the recognition rate
s achieved by independent human testers. Two methods for segmenting handwri
tten components into strokes are proposed, with better experimental results
for the method based on biological models of handwriting, in terms of cons
istency and network complexity. A systematic experimental study of differen
t codification schemes is also described, based on Shannon entropy and clus
tering maps. Finally, some steps towards the construction of an allograph l
exicon are shown, that exploit the generation of fuzzy-rules by FasArt arch
itecture.