Font Size: a A A

Research On Grammar Checking In English Essay Intelligent Marking

Posted on:2014-02-10Degree:MasterType:Thesis
Country:ChinaCandidate:L LiuFull Text:PDF
GTID:2248330398972009Subject:Computer technology
Abstract/Summary:PDF Full Text Request
It is estimated that over a billion people are speaking or learning English as their second or foreign language, and the numbers are growing day by day. English writing is an important part of learning English and is also the basic requirements for students in English examination. As we all know, writing is the weakest part in ESL’s English learning. Many ESL students need to improve writing skills to pass various language tests; thus, English teachers need to read many compositions and provide feedback. If learners can be pointed out errors timely and obtain the proposed corrections and examples, their writing skills can be improved rapidly.However, in the actual teaching work, because of the heavier burden on teachers’work and other objective reasons, resulting in teachers have not enough time and energy to read all compositions and provide feedback, which cause students have poor English writing skills. To help ESL teachers reduce their teaching load and to give students faster feedback, various English grammar checkers have been developed. Few of these PC-based grammar checkers, however, are widely available to ESL learners.This subject designs the grammar checking module of an intelligent English essay marking system. We use a combination of rule-based method and Statistical-based method, not only use rule-based method to identify grammar errors in the text, but also use statistical data as a supplement. The grammar checker uses the rule-based method to detect errors, and generates suggestions. However, there are a lot of error will not be detected, and false positives also will be produced. In order to avoid the problem of false positives, we use statistical-based methods to calculate the score of the sentence corrected (P (s)). If P (s)> T (default preset threshold), system will accept the suggestions, then output the suggestions and feedbacks; If P(s)<T, the suggestions will be abandoned. The system’s evaluation results show that our grammar checker is very effective to find the common grammar errors in the Chinese students’ English essays. And the grammar checker has been successfully integrated the well-known online English essay intelligent marking system--pigai.org, which has a low false alarm rate and is able to give feedbacks and correction suggestions timely.
Keywords/Search Tags:Grammar Checking, Rules-based, Statistical-based, N-grams, ESL Writing
PDF Full Text Request
Related items