Corpus-based online common error detector

Abstract

In this paper the theoretical underpinnings and the practical construction of a corpus-based online common error detector are described. The selection of errors for inclusion was based on analysis of published accounts of common errors in the writing of Japanese learners of English, which were confirmed through statistical corpus analysis. Harnessing the concept of regular expressions and utilizing a MySQL database, the online error detector provides instant feedback on common errors in academic discourse that standard spell and grammar checker programs do not detect.