Part of Advances in Neural Information Processing Systems 5 (NIPS 1992)
Hinrich Schütze
Representations for semantic information about words are neces(cid:173) sary for many applications of neural networks in natural language processing. This paper describes an efficient, corpus-based method for inducing distributed semantic representations for a large num(cid:173) ber of words (50,000) from lexical coccurrence statistics by means of a large-scale linear regression. The representations are success(cid:173) fully applied to word sense disambiguation using a nearest neighbor method .