Title :
Consistency of stochastic context-free grammars from probabilistic estimation based on growth transformations
Author :
Sánchez, Joan-Andreu ; Benedí, José-Miguel
Author_Institution :
Dept. de Sistemas Inf. y Comput., Univ. Politecnica de Valencia, Spain
fDate :
9/1/1997 12:00:00 AM
Abstract :
An important problem related to the probabilistic estimation of stochastic context-free grammars (SCFGs) is guaranteeing the consistency of the estimated model. This problem was considered by Booth-Thompson (1973) and Wetherell (1980) and studied by Maryanski (1974) and Chaudhuri et al. (1983) for unambiguous SCFGs only, when the probability distributions were estimated by the relative frequencies in a training sample. In this work, we extend this result by proving that the property of consistency is guaranteed for all SCFGs without restrictions, when the probability distributions are learned from the classical inside-outside and Viterbi algorithms, both of which are based on growth transformations. Other important probabilistic properties which are related to these results are also proven
Keywords :
context-free grammars; formal languages; formal specification; maximum likelihood estimation; probability; Viterbi algorithms; consistency; growth transformations; inside-outside algorithms; probabilistic estimation; probability distributions; stochastic context-free grammars; Computational linguistics; Context modeling; Frequency estimation; Natural languages; Pattern recognition; Probability distribution; Speech recognition; Stochastic processes; Viterbi algorithm;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on