Further the previous experiments the use of Weibull and Haustein functions for the approximation of the dependence between sample size and resulting vocabulary size is analyzed. The short stories by А.Т.Averchenko was chosen as the material for the experiment (total volume is more than 500 000 tokens). Haustein function is not proved to be the preferable one for the approximation of the dependency that may result from the different character of vocabulary growth for the authors under investigation.
Translated title of the contributionXIX –XX CENTURIES’ RUSSIAN SHORT STORIES CORPUS. APPROXIMATION MODELS
Original languageRussian
Title of host publicationТруды международной конференции "Корпусная лингвистка-2019"
EditorsВ.П. Захаров
Place of PublicationСПб.
PublisherИздательство Санкт-Петербургского университета
Pages379-386
StatePublished - 2019

    Research areas

  • Authors’ lexicography, statistical modeling, stylometry

ID: 43144838