time_series:dtw

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
Last revisionBoth sides next revision
time_series:dtw [2014/12/24 23:57] – phreazertime_series:dtw [2014/12/25 01:05] – [Literatur] phreazer
Line 4: Line 4:
 Iterative Deepening Dynamic Time Warping for Time Series Iterative Deepening Dynamic Time Warping for Time Series
  
 +Normalisieren der Distanzen mit der Länge des Warping Path. (Siehe "Indexing Multi-Dimensional Time-Series").
 ==== PDTW ==== ==== PDTW ====
  
Line 13: Line 14:
  
 $$d(i,j) = \sum_{k=1}^K(t(k,i)-r(k,j))^2$$ $$d(i,j) = \sum_{k=1}^K(t(k,i)-r(k,j))^2$$
 +
 +==== Multiple Multidimensional DTW ====
 +
 +Gleichzeitiges Alignment mehrerer Zeitreihen
 +
 +Quelle: Multiple Multidimensional Sequence Alignment Using Generalized Dynamic Time Warping
 ==== Literatur ==== ==== Literatur ====
 === A Scalable Method for Time Series Clustering === === A Scalable Method for Time Series Clustering ===
Line 19: Line 26:
 Dynamic Time Warping (DTW) has been applied in time series mining to resolve the difficulty caused when clustering time series of varying lengths in Euclidean space or containing possible out-of-phase similarities (Berndt & Clifford, 1994; Keogh, 2002; Ratanamahatana & Keogh, 2004). [...] but is not defined if a single data point is missing. Dynamic Time Warping (DTW) has been applied in time series mining to resolve the difficulty caused when clustering time series of varying lengths in Euclidean space or containing possible out-of-phase similarities (Berndt & Clifford, 1994; Keogh, 2002; Ratanamahatana & Keogh, 2004). [...] but is not defined if a single data point is missing.
  
 +
 +=== Indexing ===
  
  
  
  • time_series/dtw.txt
  • Last modified: 2014/12/25 01:11
  • by phreazer