A Second Language Acquisition Model Using Example Generalization and Concept Categories
Ari Rappoport and Vera Sheinman
ACL 2005 Workshop on Psycho-computational Models of Human Language Acquisition 2005.

Abstract

We present a computational model of acquiring a second language from example sentences. Our learning algorithms build a construction grammar language model, and generalize using form-based patterns and the learner’s conceptual system. We use a unique professional language learning corpus, and show that substantial reliable learning can be achieved even though the corpus is very small. The model is applied to assisting the authoring of Japanese language learning corpora.

[Edit]