6533b82cfe1ef96bd128f302
RESEARCH PRODUCT
FPCA Algorithm For Waveform Clustering
Giada AdelfioMarcello ChiodiA. D AlessandroDario Luziosubject
FPCAclustering of curveswaveformsSettore SECS-S/01 - Statisticadescription
Similar features between waveform data recorded for earthquakes at different time instants could suggest similar behavior of the source process of the corresponding source seismic process. In this paper we combine the aim of finding clusters from a set of individual waveform curves with the functional nature of data, applying a variant of a k-means algorithm based on the principal component rotation of data. This approach overcome the limitation of the cross-correlation, and represents an alternative to methods based on the interpolation of data by splines or linear fitting.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2011-01-01 |