Replicating L2 learning in a Computational Model
- Libby Barak, Rutgers University, Newark, New Jersey, United States
- Scott Cheng-Hsin Yang, Math and Computer Science, Rutgers University, Newark, New Jersey, United States
- Chirag Rank, Rutgers University, Newark, New Jersey, United States
- Patrick Shafto, Rutgers University - Newark, Newark, New Jersey, United States
AbstractHundreds of millions of people learn a second language (L2).1When learning a specific L2, there are common errors for native speakers of a given L1 language, suggesting specific effects of L1 on L2 learning. Nevertheless, language instruction materials are designed based only on L2. We develop a computational model that mimics the behavior of a non-native speaker of a specific language to provide a deeper understanding of the problem of learning a second language. We use a Naive Bayes to model prepositional choices in English (L2) by native Mandarin (L1) speakers. Our results show that both correct and incorrect responses can be explained by the learner's L1 information. Moreover, our model predicts incorrect choices with no explicit training data of non-native mistakes. Our results thus provide a new medium to analyze and develop tools for L2 teaching.