TY - JOUR
T1 - Multiscale data analysis - Information fusion and constant-time clustering
AU - Murtagh, F.
AU - Aussem, A.
AU - Starck, J. L.
PY - 1997
Y1 - 1997
N2 - 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).
AB - 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).
KW - Multiscale data analysis
KW - Information fusion
KW - Constant-time clustering
UR - http://www.scopus.com/inward/record.url?scp=33750415529&partnerID=8YFLogxK
U2 - 10.1016/S0083-6656(97)00039-1
DO - 10.1016/S0083-6656(97)00039-1
M3 - Article
AN - SCOPUS:33750415529
VL - 41
SP - 359
EP - 364
JO - New Astronomy Reviews
JF - New Astronomy Reviews
SN - 1387-6473
IS - 3
ER -