Národní úložiště šedé literatury Nalezeno 2 záznamů.  Hledání trvalo 0.01 vteřin. 
Using Explicit Semantic Analysis to Link in Multi-Lingual Document Collections
Žilka, Lukáš ; Otrusina, Lubomír (oponent) ; Smrž, Pavel (vedoucí práce)
Keeping links in quickly growing document collections up-to-date is problematic, which is exacerbated by their multi-linguality. We utilize Explicit Semantic Analysis to help identify relevant documents and links across languages without machine translation. We designed and implemented several approaches as a part of our link discovery system. Evaluation was conducted on Chinese, Czech, English and Spanish Wikipedia. Also, we discuss the evaluation methodology for such systems and assess the agreement between links on different versions of Wikipedia. In addition, we evaluate properties of Explicit Semantic Analysis which are important for its practical use.
Using Explicit Semantic Analysis to Link in Multi-Lingual Document Collections
Žilka, Lukáš ; Otrusina, Lubomír (oponent) ; Smrž, Pavel (vedoucí práce)
Keeping links in quickly growing document collections up-to-date is problematic, which is exacerbated by their multi-linguality. We utilize Explicit Semantic Analysis to help identify relevant documents and links across languages without machine translation. We designed and implemented several approaches as a part of our link discovery system. Evaluation was conducted on Chinese, Czech, English and Spanish Wikipedia. Also, we discuss the evaluation methodology for such systems and assess the agreement between links on different versions of Wikipedia. In addition, we evaluate properties of Explicit Semantic Analysis which are important for its practical use.

Chcete být upozorněni, pokud se objeví nové záznamy odpovídající tomuto dotazu?
Přihlásit se k odběru RSS.