Approach In Automatic Detection And Correction Of Errors In Chinese Text Based On Feature And Learning

Language models adopted by most existing error detection and correction approaches of Chinese text are N-Gram models of character, word or POS tag. Their deficiencies are that only local language constraint is employed and there is no language model unification process. A feature-based automatic error detection and correction approach is presented. It uses both local language features and wide-scope semantic features. Winnow is adopted in the learning step. In experiment, this method achieves error detection recall rate of 85% precise rate of 41% and error correction rate of 51%. It shows better performance than existing approaches based on N-Gram models.