Korpusové zpracování korespondenčních textů: morfologické značkování
Corpus processing of corresponding texts: problems of morphological tagging

On the issue of corpus tagging containing substandard language phenomena

Author(s): Dana Hlaváčková
Subject(s): Theoretical Linguistics, Morphology, Evaluation research
Published by: Masarykova univerzita nakladatelství
Keywords: private correspondence; corpus; lemmatization; morphological tagging; disambiguation;
Summary/Abstract: This article summarizes the experience with the corpus processing of the corresponding texts. Attention is paid mainly to lemmatization, morphological tagging and disambiguation of texts with a high frequency of substandard linguistic phenomena. The procedure for necessary adjustments of morphological analyzer, the proportion of manual editing and the results obtained are specified.

• Details
• Contents