作者: Iryna A. Voloshynovska
DOI: 10.1080/09296174.2011.583405
关键词:
摘要: Abstract In the present work, an optimal function for approximation of word probability distribution in natural language is tentatively sought. An analysis Zipf's rank-frequency and its modifications performed. The original Lavalette's law modified to get most accurate language. suggested modification has been tested on a broad variety belletristic scientific texts. Implication text features fitting parameters suggested. conclusions are made general tendencies reasons possible deviations discussed.