Combining trigram and automatic weight distribution in Chinese spelling error correction

Abstract

The researches on spelling correction aimmg at detecting errors in texts tend context-sensitive spellng error correction, which is more difficult than traditional isloated-word error correction. A novel and efficient algorithm for the system of Chinese spelling error correction, CInsunSpell, is presented. In this system, the work of correction lncludes two parts: checking phase and correcting phase. At the first phase, a Trigram algorithm within one fixed-size window is designed to locate potenuat errors in local area. This second phase employs a new method ot automatically and dynamically distributing weights among the characters in the confusion set as well as in the Bayesian language model. The tactics used abov exhibits good performances.