Random Projections for Dimension Reduction in Information Retrieval Systems

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F03%3A00008379" target="_blank" >RIV/61989100:27240/03:00008379 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Random Projections for Dimension Reduction in Information Retrieval Systems
Original language description
Information retrieval deals with storage and searching of textual documents. One from the models of the information retrieval systems is the vector model. In vector model the term-by-document matrix is formed from vectors of documents. Some methods of the term-by-document matrix reduction have been developed in the past. Random projection is a method of dimensionality reduction significantly faster than principal component analysis and singular value decomposition. It uses projection matrix with zero-mean unit-variance random distribution and its error can be controlled easily. Tests on image data were executed recently to show distortion caused by projection. In this paper we present a measurement of precision and recall in vector model performed on real-life text data by comparison of results before and after the projection.
Czech name
—
Czech description
—

Project
<a href="/en/project/GA101%2F02%2F0072" target="_blank" >GA101/02/0072: Analysis and solution of selected nonlinear problems of elasticity by the finite element method</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2003
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)