Improving CoGrOO: the Brazilian Portuguese Grammar Checker
William D. Colen M. Silva, Marcelo Finger · 2013
Abstract. This paper highlights the main results obtained in an effort to improve the grammar checker CoGrOO, a hybrid system which initially annotates the text using statistical Natural Language Processing (NLP) techniques, and then apply a rule-based analysis to identify possible grammar errors. The goal was to reduce omissions and false alarms while improving true positives without adding new error rules. The work contributed with a detailed evaluation of low-level NLP modules and the results were compared to state-of-the-art results. The low-level NLP modules are available as open source software, thus improvements on their effectiveness will make them robust, free and ready-to-use alternatives for other systems. 1.