In this paper, we present a thorough testing of various LNRE (Large Number or Rare Events) models for the prediction of vocabulary size and other quantities at sizes larger than the sample size. The main conclusion is that none of the models appears to be empirically adequate. Preliminary evidence suggests that this might be due to non-randomness effects.
Testing the extrapolation quality of word frequency models
BARONI, MARCO;
2005
Abstract
In this paper, we present a thorough testing of various LNRE (Large Number or Rare Events) models for the prediction of vocabulary size and other quantities at sizes larger than the sample size. The main conclusion is that none of the models appears to be empirically adequate. Preliminary evidence suggests that this might be due to non-randomness effects.File in questo prodotto:
Eventuali allegati, non sono esposti
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.