Usage-Based Learning of Grammatical Constructions
Speaker: Nancy Chang
Institution: International Computer Science Institute
Abstract:
In this talk I explore and formalize the view that grammar learning is driven by meaningful language use in context. On this view, the goal of a language learner is to become a better language user --- in particular, by acquiring linguistic constructions (structured mappings between form and meaning) that facilitate successful communication. I present a computational model in which all aspects of the language learning problem are reformulated in line with these assumptions. The representational basis of the model is a construction-based grammar formalism suitable for capturing constituent structure and relational constraints. This formalism plays a central role in two processes: language understanding, which uses constructions to interpret utterances in context; and language learning, which seeks to improve comprehension by making judicious changes to the current grammar. The resulting integrated model of language structure, use and acquisition offers a more faithful approximation of the child's learning situation than many traditional approaches; it also provides a concrete realization of a broad array of proposals from constructivist and constructional approaches to language acquisition and use. When exposed to a corpus of child-directed utterances annotated with situation descriptions, the model gradually acquires the concrete word combinations, item-based constructions and simple grammatical markers that constitute the first steps toward adult language.
