Automatic Acquisition of Hyponyms from Large Text Corpora-英文文献.pdf

Automatic Acquisition of Hyponyms ~om Large Text Corpora M a r t i A. H e a r s t C o m p u t e r Science Division, 571 E v a n s Hall U n i v e r s i t y of California, Berkeley Berkeley, CA 94720 and X e r o x Palo A l t o R e s e a r c h C e n t e r m a r t i ~ c s , berkeley, edu A b s t r a c t following sentence: 1. We describe a method for the automatic acquisition (SI) The bow l u t e , such as the Bambara ndang, of the hyponymy lexical relation from unrestricted is plucked and has an i n d i v i d u a l text. Two goals motivate the approach: (i) avoid- curved neck :for each string. ance of the need for pre-encoded knowledge and (ii) applicability across a wide range of text. We identify Most fluent readers of English who }lave never be- a set of lexico-syntactic patterns that are easily rec- fore encountered the term q3amhara ndang will nev- ognizable, that occur iYequently and across text genre ertheless from this sentence infer that a Bambara boundaries, and that indisputably indicate the lexical udang is a kind of bow Iute. This is true even if relation of interest. We describe a method for discov- tile reader has only a fuzzy conce