Conserving Fuel in Statistical Language Learning: Predicting Data Requirements

dc.creatorLauer, Mark
dc.date1995-09-07
dc.date.accessioned2026-07-07T09:10:00Z
dc.date.available2026-07-07T09:10:00Z
dc.descriptionIn this paper I address the practical concern of predicting how much training data is sufficient for a statistical language learning system. First, I briefly review earlier results and show how these can be combined to bound the expected accuracy of a mode-based learner as a function of the volume of training data. I then develop a more accurate estimate of the expected accuracy function under the assumption that inputs are uniformly distributed. Since this estimate is expensive to compute, I also give a close but cheaply computable approximation to it. Finally, I report on a series of simulations exploring the effects of inputs that are not uniformly distributed. Although these results are based on simplistic assumptions, they are a tentative step toward a useful theory of data requirements for SLL systems.
dc.description8 pages
dc.identifierhttps://arxiv.org/abs/cmp-lg/9509002
dc.identifierhttp://arxiv.org/abs/cmp-lg/9509002
dc.identifierEighth Australian Joint Conference on Artificial Intelligence, Canberra, 1995.
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/151235
dc.subjectComputation and Language
dc.titleConserving Fuel in Statistical Language Learning: Predicting Data Requirements
dc.typetext

Files

Collections