Pregled bibliografske jedinice broj: 728713
Toward Network-based Keyword Extraction from Multitopic Web Documents
Toward Network-based Keyword Extraction from Multitopic Web Documents // Proceedings of the 6th International Conference on Information Technologies and Information Society (ITIS 2014) / Levnajić, Zoran ; Boshkoska, Biljana Mileva (ur.).
Novo Mesto: Fakulteta za informacijske študije v Novem mestu, 2014. str. 18-27 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 728713 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Toward Network-based Keyword Extraction from Multitopic Web Documents
Autori
Šišović, Sabina ; Martinčić-Ipšić, Sanda ; Meštrović, Ana
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the 6th International Conference on Information Technologies and Information Society (ITIS 2014)
/ Levnajić, Zoran ; Boshkoska, Biljana Mileva - Novo Mesto : Fakulteta za informacijske študije v Novem mestu, 2014, 18-27
ISBN
978-961-93391-3-8
Skup
International Conference on Information Technologies and Information Society (ITIS2014)
Mjesto i datum
Šmarješke toplice, Slovenija, 05.11.2014. - 07.11.2014
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
keyword extraction; complex networks; co-occurrence language networks; Croatian texts; selectivity
Sažetak
In this paper we analyse the selectivity measure calculated from the complex network in the task of the automatic keyword extraction. Texts, collected from different web sources (portals, forums), are represented as directed and weighted co-occurrence complex networks of words. Words are nodes and links are established between two nodes if they are directly co-occurring within the sentence. We test different centrality measures for ranking nodes – keyword candidates. The promising results are achieved using the selectivity measure. Then we propose an approach which enables extracting word pairs according to the values of the in/out selectivity and weight measures combined with filtering.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo, Informacijske i komunikacijske znanosti
POVEZANOST RADA
Projekti:
Uniri-LangNet
Ustanove:
Fakultet informatike i digitalnih tehnologija, Rijeka