Pregled bibliografske jedinice broj: 486603
Robust Keyphrase Extraction For A Large-Scale Croatian News Production System
Robust Keyphrase Extraction For A Large-Scale Croatian News Production System // Proceedings of the Seventh International Conference on Formal Approaches to South Slavic and Balkan Languages / Tadić, Marko ; Dimitrova-Vulchanova, Mila ; Koeva, Svetla (ur.).
Zagreb: Hrvatsko društvo za jezične tehnologije, 2010. str. 59-66 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 486603 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Robust Keyphrase Extraction For A Large-Scale Croatian News Production System
Autori
Mijić, Jure ; Šnajder, Jan ; Dalbelo Bašić, Bojana
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the Seventh International Conference on Formal Approaches to South Slavic and Balkan Languages
/ Tadić, Marko ; Dimitrova-Vulchanova, Mila ; Koeva, Svetla - Zagreb : Hrvatsko društvo za jezične tehnologije, 2010, 59-66
ISBN
978-953-55375-2-6
Skup
Seventh International Conference on Formal Approaches to South Slavic and Balkan Languages
Mjesto i datum
Dubrovnik, Hrvatska, 04.10.2010. - 06.10.2010
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
keyphrase extraction; news production system; information extraction; natural language processing; Croatian language
(keyword extraction; news production system; information extraction; natural language processing; Croatian language)
Sažetak
Summarizing an article with just a few keyphrases can be a difficult task, even for trained experts. Large-scale keyphrase extraction requires a method that is fast and reliable, and yet relatively effective. In this paper we describe such a keyphrase extraction system developed for a large- scale Croatian news production system. We describe how the system works and evaluate the implemented keyphrase extraction methods using a gold set annotated by human annotators. The results indicate that, despite the simplicity of our approach, the performance of the system is comparable to that of the human annotators.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Projekti:
036-1300646-1986 - Otkrivanje znanja u tekstnim podacima (Dalbelo-Bašić, Bojana, MZO ) ( CroRIS)
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb