Naslov
Concept Decomposition by Fuzzy k-means Algorithm

Autori
Dobša, Jasminka ; Dalbelo Bašić, Bojana

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

Izvornik
Proceedings of the IEEE/WIC International Conference on Web Intelligence, WI 2003 / Liu, Jiming ; Liu, Liu, Chunnian ; Klusch, Matthias ; Zhong, Ning ; Cercone, Nick - : IEEE Computer Society, Washington, Brussels, Tokyo, 2003, 684-688

ISBN
9780769519326

Skup
IEEE/WIC International Conference on Web Intelligence (WI 2003)

Mjesto i datum
Halifax, Kanada, 13.10.2003. - 17.10.2003

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
information retrieval ; Latent Semantic Indexing ; concept decomposition ; singular value decomposition ; fuzzy k-means algorithm

Sažetak
The method of Latent semantic indexing (LSI) is information retrieval technique which uses a low-rank singular value decomposition (SVD) of the term-document matrix. Although LSI method has empirical success, it suffers from lack of interpretation for the low-rank approximation and consequently a lack of controls for accomplishing specific tasks in information retrieval. A method introduced by Dhillon and Modha is improvement in that direction. It uses centroids of clusters or so called concept decomposition for lowering the rank of term-document matrix. Our work is focused on improvements of that method using fuzzy k-means algorithm . Also, we compare precision of information retrieval for the two methods mentioned above.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA