
In Czech: Modifikace bayesovského disambiguátoru
Modification of the Bayes Disambiguator
This paper deals with bayessian disambiguation. It proposes some heuristic improvements. According to our preliminary tests, the precision is in between 50% - 90% depending on thematic similarity of the classes in the testing corpora.
Keywords: disambiguation, word sense, EuroWordNet, NLP
Year: 2005
Download:
Full text [90 kB]

Authors of this publication:

Michal Toman
E-mail: mtoman@kiv.zcu.cz
Michal graduated at UWB in 2003, specialized in software engineering. Currently, he is a PhD student interested in information retrieval, multilingual text processing, word sense disambiguation and knowledge discovery.

Karel Ježek
Phone: +420 377632475, 377632400
E-mail: jezek_ka@kiv.zcu.cz
WWW: http://www-kiv.zcu.cz/~jezek_ka/
Karel is a group coordinator and a supervisor of PhD students working at research projects of this Group.