Deducing linguistic structure from the statistics of large corpora
Eric Brill, David M. Magerman, Mitchell P. Marcus, Beatrice Santorini · 1990
Within the last two years, approaches using both stochastic and symbolic techniques have proved adequate to deduce lexical ambiguity resolution rules with less than 3-4% error rate, when trained on moderate sized (500K word) corpora of English text (e.g. Church, 1988; Hindle, 1989). The success of these techniques suggests that much of the grammatical structure of language may be derived automatically through distributional analysis, an approach attempted and abandoned in the 1950s.