Cursive script recognition using wildcards and multiple experts

Variability in handwriting styles suggests that many letter recognition eng ines cannot correctly identify some handwritten letters of poor quality at reasonable computational cost. Methods that are capable oi searching the re sulting sparse graph of letter candidates are therefore required. The metho d presented here employs 'wildcards' to represent missing letter candidates . Multiple experts are used to represent different aspects of handwriting. Each expert evaluates closeness of match and indicates its confidence. Expl anation experts determine the degree to which the word alternative under co nsideration explains extraneous Letter candidates. Schemata for normalisati on and combination of scores are investigated and their performance compare d. Hill climbing yields near-optimal combination weights that outperform co mparable methods on identical dynamic handwriting data.