Zobrazit v EDS

Concept Decomposition by Fuzzy k-means Algorithm

Uloženo v:
Podrobná bibliografie
Název: Concept Decomposition by Fuzzy k-means Algorithm
Autoři: Dobša, Jasminka, Dalbelo Bašić, Bojana
Přispěvatelé: Liu, Jiming, Liu, Liu, Chunnian, Klusch, Matthias, Zhong, Ning, Cercone, Nick
Informace o vydavateli: 2003.
Rok vydání: 2003
Témata: concept decomposition, singular value decomposition, information retrieval, fuzzy k-means algorithm, Latent Semantic Indexing
Popis: 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.
Druh dokumentu: Conference object
Jazyk: English
Přístupová URL adresa: https://www.bib.irb.hr/144662
Přístupové číslo: edsair.dedup.wf.002..9f2c7968f5f503ed9f41f1c8fb789b1f
Databáze: OpenAIRE
Popis
Abstrakt: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.