• DocumentCode
    1050125
  • Title

    On the bias of the Turing-Good estimate of probabilities

  • Author

    Juang, B.H. ; Lo, S.H.

  • Author_Institution
    AT&T Bell Labs., Murray Hill, NJ, USA
  • Volume
    42
  • Issue
    2
  • fYear
    1994
  • fDate
    2/1/1994 12:00:00 AM
  • Firstpage
    496
  • Lastpage
    498
  • Abstract
    Good´s (1953) estimate, based on Turing´s formula, was suggested for estimating the probabilities of words in text as well as of species in a mixed population and was found particularly useful for the probability of unseen classes. The authors address the issue of bias in Good´s estimate and propose an alternative to reduce this bias. This may be important in the construction of a language model for speech recognition where sparse data and low probability events are key problems
  • Keywords
    estimation theory; probability; speech recognition; Turing-Good probability estimate; bias; language model; low probability events; mixed population; sparse data; species; speech recognition; text; words; Bayesian methods; Maximum likelihood estimation; Natural languages; Probability; Speech recognition; Text analysis; Tin; Vocabulary;
  • fLanguage
    English
  • Journal_Title
    Signal Processing, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    1053-587X
  • Type

    jour

  • DOI
    10.1109/78.275640
  • Filename
    275640