Mathematical model of character recognition

This article describes a mathematical model of character recognition for interpretation of historical handwritten documents. Historical shorthand records of the XIX century are taken as objects. The article describes a mathematical model based on the Bayesian approach. A method for estimating the accuracy of character recognition, algorithm solutions, and labor-intensive task of building high dimensional matrices to calculate the estimates of the probability of entering fragments in to the test is described. The results of estimating the accuracy of recognition for the training sample and estimating the probability of occurrence of the five most frequently used words in the test are given. An algorithm to implement the proposed mathematical model of the information system for the recognition of handwritten historical documents is described.