Naslov
A general framework for mining relations for the semantic web

Autori
Doko, Alen ; Štula, Maja

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
IIWeb '12 Proceedings of the Ninth International Workshop on Information Integration on the Web / Ullas Nambiar ; Zaiqing Nie - New York (NY) : The Association for Computing Machinery (ACM), 2012, 1-5

ISBN
978-1-4503-1239-4

Skup
The Ninth International Workshop on Information Integration on the Web

Mjesto i datum
Scottsdale (AZ), Sjedinjene Američke Države, 20.03.2012. - 23.03.2012

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Semantic web; Association rules; Ontology learning

Sažetak
Automatic ontology learning is inevitable for the expansion of the Semantic Web. This paper lists basic problems of ontology learning at first. Then it proposes a general framework for mining relations from various sources to build ontologies for the semantic web. The framework aims at solving the problems of ontology learning by concentrating on the generic relation "is related to" between objects on the web. It is based on the fact that every data source on the web has a certain clearly defined structure which can be used to automatically extract semantic relations. To define the strength of the semantic relation frequent item set mining is proposed.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA