Multiscale data analysis - Information fusion and constant-time clustering

F. Murtagh, A. Aussem, J. L. Starck

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We describe the use of the wavelet transform for multivariate data analysis problems. In prediction, a multiscale transform of time-varying data can allow forecasts of each scale, followed by a combining of the individual forecasts. The use of a wavelet transform with noise modeling for point pattern clustering can lead to the result, which initially appears counter-intuitive, of clustering in constant computational time, O(1).

Original languageEnglish
Pages (from-to)359-364
Number of pages6
JournalVistas in Astronomy
Volume41
Issue number3
DOIs
Publication statusPublished - 1997
Externally publishedYes

    Fingerprint

Cite this