Extension matrix algorithm for multi-class problem with overlay area

In order to deal with the multi-class problems with overlay, the improved extension matrix algorithm was proposed. The heuristic search based on average entropy was used to get the approximate solutions of the shortest equation. The potential function was constructed to estimate the probability density function of the overlay area between positive and negative examples, to generate recognition rules of overlay area, so the nonlinear discriminatory interface of the interclass areas may be obtained. The algorithms have been applied to the hand-written Chinese character recognition and the result has shown that the improved algorithm is effective.