Analysis of an Associative Text Recognition Method

摘要

An associative text recognition method is thoroughly analyzed in the paper. The method finds in a large N-gram network near-optimal paths that match well with the erroneous text or sentences used as an input. The effects of various factors, including error probability per character, N-gram length, misleading of search, database similarity, etc., are analyzed. Optimal suggestions for the parameters of the method are also given.