Design And Implementation Of English Grammar Checker Assist Reviewers Writing System

With the development of technology and algorithms for syntax checking, gram-mar checking systems are widely used in various fields. English composition auxiliary review system is a useful tool and a system to help and assist reviewers for quick and efficient evaluation to the level of English composition. The system includes a spelling checker, grammar checker, writing management, scoring system, and many other mod-ules, in which grammar checker is the key and difficult point of the system’s designa-tion. At present, syntax checking systems mostly adopt single-mode design, which has the problem of low accuracy rate and poor effect in English writing grammar check-ing. The main contents of this paper to provide the method to improve the grammar checking algorithms so as to raise its accuracy, design and develop a grammar checker subsystem.To study and solve these problems, this article is from the following three aspects, design and implement a suitable English writing auxiliary marking syntax checking system. Syntax checking system uses multi-module design, including a rule-based syn-tax checking modules and machine learning algorithms based syntax checking modules. First, the grammar checker grammar checker module design based on multi-rule, multi-layered database for common grammar rules of English grammar provide error correc-tion capabilities. To improve the grammar checker module syntax checking accuracy, based on the statistical system into grammar model, the use of N-gram model to find out the rules of grammar errors evaluated. The second, directed to articles and prepositions ineffective correction problem, design a module independent grammar of articles and prepositions. Designed with syntax checking model based on machine learning algo-rithms, the syntax checking issues of equivalent articles and prepositions classification problem, troubleshoot problems specific grammar correction. For low recall preposi-tion model in use, resulting balance of positive and negative samples, this paper presents a wrong interpolation algorithm, to reduce reliance on the original sentence preposition model features, experimental verification model to improve the recall, while ensuring accurate rate basically stable. Third, the system was designed and implemented gram-mar checking system, provide external interfaces through web services.Experimental results show that the grammar check system for domestic students’ English writing sample syntax checking, the overall result was39.07F value. Good system performance, using asynchronous mode, you can also check on the hundreds of statements, to the purpose of the design.