%0 Journal Article
%T Combining Statistical and Rule-Based Approaches to Morphological Tagging of Czech Texts
%A Drahom¨ªra "johanka" Spoustov¨¢
%J The Prague Bulletin of Mathematical Linguistics
%@ 1804-0462
%D 2008
%I 
%R 10.2478/v10108-009-0002-x
%X This article is an extract of the PhD thesis (Spoustov¨¢, 2007) and it extends the article (Spoustov¨¢ et al., 2007). Several hybrid disambiguation methods are described which combine the strength of hand-written disambiguation rules and statistical taggers. Three different statistical taggers (HMM, Maximum-Entropy and Averaged Perceptron) and a large set of hand-written rules are used in a tagging experiment using Prague Dependency Treebank. The results of the hybrid system are better than any other method tried for Czech tagging so far.
%U http://versita.metapress.com/content/3257651658423687/?p=9834849923904c7cabe3f5c258413e32&pi=1