Correcting Grammatical Verb Errors

Abstract

Verb errors are some of the most common mistakes made by non-native writers of English but some of the least studied. The reason is that dealing with verb errors requires a new paradigm; essentially all research done on correcting grammatical errors assumes a closed set of triggers -e.g., correcting the use of prepositions or articles -but identifying mistakes in verbs necessitates identifying potentially ambiguous triggers first, and then determining the type of mistake made and correcting it. Moreover, once the verb is identified, modeling verb errors is challenging because verbs fulfill many grammatical functions, resulting in a variety of mistakes. Consequently, the little earlier work done on verb errors assumed that the error type is known in advance.We propose a linguistically-motivated approach to verb error correction that makes use of the notion of verb finiteness to identify triggers and types of mistakes, before using a statistical machine learning approach to correct these mistakes. We show that the linguistically-informed model significantly improves the accuracy of the verb correction approach.

Neural Grammatical Error Correction Systems with Unsupervised Pre-training on Synthetic Data

“…In a naïve approach, words can be replaced randomly within a vocabulary, but this may result in unrealistic error patterns that do not resemble those observed in the genuine data. More accurate errors can be generated by replacing words only within confusion sets if such a confusion set consists of words that are commonly confused with each other (Rozovskaya and Roth, 2010;Rozovskaya et al, 2014;Bryant and Briscoe, 2018).…”

The Illinois-Columbia System in the CoNLL-2014 Shared Task

“…In addition, plural and singular forms are added for all words tagged as nouns, and inflectional forms are added for words tagged as verbs. For more detail on correcting verb errors, we refer the reader to Rozovskaya et al (2014).…”

“…The AP classifiers all make use of richer sets of features than the native-trained classifiers: the article, noun number, and preposition classifiers employ features that use POS information, while the verb agreement classifier also makes use of dependency features extracted using a parser (de Marneffe et al, 2008). For more detail on the features used in the agreement module, we refer the reader to Rozovskaya et al (2014). Finally, all of the AP models use the source word of the author as a feature and, similar to the article AP classifier (Section 3), implement the error inflation method.…”

Using Learner Data to Improve Error Correction in Adjective–Noun Combinations

“…More recently, however, the need to address other error types has been recognised (Kochmar and Briscoe, 2014;Ng et al, 2014;Rozovskaya et al, 2014;Sawai et al, 2013;Dahlmeier and Ng, 2011). Among these, errors in content words are the third most frequent error type after errors in articles and prepositions (Leacock et al, 2014;Ng et al, 2014).…”