Grammar Automatic Checking System for English Abstract of Master's Thesis

Although the technology of natural language processing was proposed several years ago, there are few Internet-based systems of English automatic grammar checking which is of help for beginner academic writers. This paper introduces our research results of an automatic grammar checking system for English article abstracts. We proposed four improved basic methods of natural language processing like sentence segmentation, multi-level indexes of words, Penn treebank and increment artificial rules for the system. Experimental results indicate that the proposed system is able to detect more grammar mistakes than other popular similar systems such as 1Checkerm Microsoft word and NonPlus. Furthermore, our system provides free service for Internet users.