A Maximum Entropy Approach to Chinese Spelling Check

Spelling check identifies incorrect writing words in documents. For the reason of input methods, Chinese spelling check is much different from English and it is still a challenging work. For the past decade years, most of the methods in detecting errors in documents are lexicon-based or probability-based, and much progress are made. In this paper, we propose a new method in Chinese spelling check by using maximum entropy (ME). Experiment shows that by importing a large raw corpus, maximum entropy can build a well-trained model to detect spelling errors in Chinese documents.