Dimensionality reduction by random mapping: fast similarity computation for clustering
- 27 November 2002
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 413-418
- https://doi.org/10.1109/ijcnn.1998.682302
Abstract
When the data vectors are high-dimensional it is computationallyinfeasible to use data analysis or patternrecognition algorithms which repeatedly compute similaritiesor distances in the original data space. It istherefore necessary to reduce the dimensionality before,for example, clustering the data. If the dimensionalityis very high, like in the WEBSOM method which organizestextual document collections on a Self-OrganizingMap, then even the commonly used dimensionality reduction...Keywords
This publication has 3 references indexed in Scilit:
- Self-Organizing MapsPublished by Springer Nature ,1995
- Indexing by latent semantic analysisJournal of the American Society for Information Science, 1990
- Self-organizing semantic mapsBiological Cybernetics, 1989