Combine Trigram and Automatic Weight Distribution in Chinese Spelling Error Correction

Abstract. The researches on spelling correction aiming at fixing the errors in context tend to focus on the context-sensitive spelling error correction, which is more difficult than the traditional checker-isolated-word error correction. A novel and efficient algorithm for the system of Chinese spelling error correction, CInsunSpell, is presented here. In this system, the work of correction sets apart into two parts: checking phase and correcting phase. At the first phase, a Trigram algorithm within one fixed-size window is designed to locate the suspicious errors in local area. The second phase employs a new thought of automatically and dynamically distributing weights among the characters in confusion set as well as the Bayesian language model. The tactics used above have exhibited good performances