Online Handwriting Recognition Using Time -Order of Lexical and Signal Co-Occurrences

Letter-polygram based Symbolic Indirect Correlation is a new method that offers significant advantages for ordered unsegmented signals. However, its application to on -line, cursive handwriting requires solving several difficult problems. (1) Reference strings of words must satisfy certa in uniformity properties on their lexical match with the lexicon of unknown words. (2) The Viterbi algorithm must be modified and trained to extract polygram -length feature matches from both the reference string and the query. (3) Efficient permutation matching algorithms are required for the comparison of lexical and feature permutations. We present our analyses, simulations and current status, and solicit suggestions to overcome these problems.