Punctuation in a Lexicalized Grammar
Christine Doran · 2000
Just as people make use of information from punctuation to structure and understand text, NLP systems can use information from punctuation in processing texts automatically. The aim of the research presented here was to explore the feasibility of treating a sizable core of punctuation phenomena at the level of the sentence grammar. A large set of punctuation rules were manually derived from naturally occurring data, and added to the XTAG English grammar. Our results confirm that punctuation can be used in analyzing sentences to increase the coverage of the grammar, reduce the ambiguity of certain word sequences and facilitate later processing of larger text units, without either adversely impacting the existing grammar or deriving analyses which would be incompatible with later levels of processing.