A Language Learning System with Automatic Feedback: An Application Based on a English-Chinese Parallel Corpus

Abstract

The learners of English as a second language generally need many practices on writing English, identification of the mistakes in their English, and feedback hints on how to correct their mistakes. A computer-assisted online system is designed to address these issues in a context of learning from a corpus of parallel English-Chinese corpus of New York Times news articles. In the system, students match chunks of English words to chunks of Chinese strings and order chunks of English to produce the original English sentences. Results from an empirical experiment indicated that the students found the tasks challenging and felt that the learning activities were meaningful in improving their knowledge of English. Their score improvement increased when they practiced more.