Using Mostly Native Data to Correct Errors in Learners' Writing
نویسنده
چکیده
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.
منابع مشابه
Native Language Interference in Writing: A case study of Thai EFL learners
AbstractThe interference of the native language in acquiring a foreign language is unavoidable. In an attempt to explore the phenomenon why this occurs, the study was conducted in English as a foreign language writing. The study also investigated how the native language interference occurred in the writing process. In fact, this qualitative study explored the reasons and the process of na...
متن کاملNative Language Interference in Writing: A case study of Thai EFL learners
AbstractThe interference of the native language in acquiring a foreign language is unavoidable. In an attempt to explore the phenomenon why this occurs, the study was conducted in English as a foreign language writing. The study also investigated how the native language interference occurred in the writing process. In fact, this qualitative study explored the reasons and the process of na...
متن کاملThe Effect of Written Corrective Feedback on EFL Students\' Writing Accuracy
The present study explores the effect of providing different types of written corrective feedback (WCF) on 79 Iranian EFL learners’ writings over time through a pre-test, immediate post-test and delayed post-test. In this way, the study first aims to investigate the extent to which different types of WCF result in improved accuracy in writing over a period of two months, and then maintain...
متن کاملUser Input and Interactions on Microsoft Research ESL Assistant
ESL Assistant is a prototype web-based writing-assistance tool that is being developed for English Language Learners. The system focuses on types of errors that are typically made by non-native writers of American English. A freely-available prototype was deployed in June 2008. User data from this system are manually evaluated to identify writing domain and measure system accuracy. Combining th...
متن کاملCorrecting ESL Errors Using Phrasal SMT Techniques
This paper presents a pilot study of the use of phrasal Statistical Machine Translation (SMT) techniques to identify and correct writing errors made by learners of English as a Second Language (ESL). Using examples of mass noun errors found in the Chinese Learner Error Corpus (CLEC) to guide creation of an engineered training set, we show that application of the SMT paradigm can capture errors ...
متن کاملThe Comparative Effect of Direct-only Correction and Direct Metalinguistic Correction on the Improvement of EFL Learners’ Writing Ability
This study aimed at comparing the impact of two types of teacher feedback on Iranian EFL learners’ writing ability and their verb tense consistency in L2 writing. Following the administration of a standardized language proficiency test (Preliminary English Test), sixty-two participants were selected and were randomly assigned to the control (direct-only correction) and experimental (direct meta...
متن کامل