Parser Adaptation to the Biomedical Domain without Re-Training

Abstract

We present a distributional approach to the problem of inducing parameters for unseen words in probabilistic parsers. Our KNN-based algorithm uses distributional similarity over an unlabelled corpus to match unseen words to the most similar seen words, and can induce parameters for those unseen words without retraining the parser. We apply this to domain adaptation for three different parsers that employ fine-grained syntactic categories, which allows us to focus on modifying the lexicon, while leaving the structure of the parser itself intact. We demonstrate uplifts for dependency recovery of 2%-6% on novel vocabulary in biomedical text.