Research on proofreading system of judicial documents based on N-gram model

Based on the characteristics of Chinese word-writing, the study research and enumerate 6 major areas prone to misuses: words, sentences, punctuation, structure, legal citation and judicial terminology. With N-gram model, the paper has developed instruments proofreading judgments model, which come up with a trusted binary collocation database by training specific judicial corpus. This database along with other resources constitutes proofreading standard databases, which become the foundation of proofreading. It has designed and implemented an intelligent proofreading system of judicial documents. The system is easy to use and expandable. It is proved that the system plays a positive role in improving the quality of judicial documents and the credibility of Justice as well.