Using Mostly Native Data to Correct Errors in Learners' Writing: A Meta-Classifier Approach

We present results from a range of experiments on article and preposition error correction for non-native speakers of English. We first compare a language model and errorspecific classifiers (all trained on large English corpora) with respect to their performance in error detection and correction. We then combine the language model and the classifiers in a meta-classification approach by combining evidence from the classifiers and the language model as input features to the metaclassifier. The meta-classifier in turn is trained on error-annotated learner data, optimizing the error detection and correction performance on this domain. The meta-classification approach results in substantial gains over the classifieronly and language-model-only scenario. Since the meta-classifier requires error-annotated data for training, we investigate how much training data is needed to improve results over the baseline of not using a meta-classifier. All evaluations are conducted on a large errorannotated corpus of learner English.