Addition of Documents' Representations in the Latent Semantic Space (CROSBI ID 546308)
Prilog sa skupa u zborniku | sažetak izlaganja sa skupa | međunarodna recenzija
Podaci o odgovornosti
Dobša, Jasminka
engleski
Addition of Documents' Representations in the Latent Semantic Space
Latent semantic indexing (LSI) is the most popular method of dimensionality reduction of original representation of textual documents in the vector space model. Collections of documents very often are dinamical because new documents constantly are added to collection. Vectors on which the projection is done in the process of dimensionality reduction are constructed on the basis of representations of all documents in the collection, and computation of the new representations in the space of reduced dimension demands recomputation of singular value decomposition. In order to overcome that problem Barry and coworkers (1995) sugessted approximative representation of added documents by projections on existing left singular vectors. They also propose method for approximative representation of added index terms. Here will be proposed modification of approximative representations of terms and documents by combination of these two methods. It is shown that representation of documents by extended list of index terms does not improve performance of information retrieval significantly.
information retrieval; latent semantic indexing; addition of documents
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
67-67.
2008.
objavljeno
Podaci o matičnoj publikaciji
Proceedings of International Conference Applied Statistics 2008
Lusa, Lara ; Stare, Janez
Ljubljana: Statistical Society of Slovenia
Podaci o skupu
International Conference Applied Statistics 2008
predavanje
21.09.2008-24.09.2008
Ribno, Slovenija