POSTECH Grammatical Error Correction System in the CoNLL-2014 Shared Task

This paper describes the POSTECH grammatical error correction system. Various methods are proposed to correct errors such as rule-based, probability n-gram vector approaches and router-based approach. Google N-gram count corpus is used mainly as the correction resource. Correction candidates are extracted from NUCLE training data and each candidate is evaluated with development data to extract high precision rules and n-gram frames. Out of 13 participating teams, our system is ranked 4 on both the original and revised annotation.