Benchmark id: inv1

Multi-Dimensional Scaling (MDS) analysis

Select Solutions for the comparisions (atleast 3, default: none for 'all')

(default: none, also if this is not one of selected Solutions)                    





Ranking with respect to the reference
Excellent Good Fair Poor
  gallovic2 (5)
  somala1 (12)
  SIVdata (1)
  causse (2)
  fsg (3)
  gallovic (4)
  gallovic3 (6)
  hobyy1 (8)
  hobyy3 (9)
  navid (10)
  hobyt (7)
  somala (11)

Dissimilarity matrix
Solution SIVdata causse fsg gallovic gallovic2 gallovic3 hobyt hobyy1 hobyy3 navid somala somala1
  SIVdata 0.00   26.92   2.90   19.77   7.85   23.64   24.27   21.54   19.75   6.94   29.61   11.69  
  causse 26.92   0.00   26.57   23.77   18.62   17.64   37.66   35.30   27.97   23.08   32.64   23.21  
  fsg 2.90   26.57   0.00   19.51   5.80   20.14   24.58   16.73   15.42   2.88   32.36   8.88  
  gallovic 19.77   23.77   19.51   0.00   11.09   8.84   43.24   23.98   32.00   19.77   35.05   18.26  
  gallovic2 7.85   18.62   5.80   11.09   0.00   10.74   32.10   19.82   17.63   7.32   25.48   9.79  
  gallovic3 23.64   17.64   20.14   8.84   10.74   0.00   47.15   19.59   26.97   19.65   30.33   16.58  
  hobyt 24.27   37.66   24.58   43.24   32.10   47.15   0.00   37.10   34.32   25.18   52.99   28.77  
  hobyy1 21.54   35.30   16.73   23.98   19.82   19.59   37.10   0.00   30.94   15.09   45.51   18.63  
  hobyy3 19.75   27.97   15.42   32.00   17.63   26.97   34.32   30.94   0.00   11.38   42.19   20.18  
  navid 6.94   23.08   2.88   19.77   7.32   19.65   25.18   15.09   11.38   0.00   38.40   10.33  
  somala 29.61   32.64   32.36   35.05   25.48   30.33   52.99   45.51   42.19   38.40   0.00   24.47  
  somala1 11.69   23.21   8.88   18.26   9.79   16.58   28.77   18.63   20.18   10.33   24.47   0.00  
Each element represents a measure of dissimilarity between pairs of models obtained using normalized square metric




Note
This analysis employs multidimensional scaling (Borg and Groenen, 2005) to compare the slip models. The dissimilarities are obtained for all pairs of slip models, viz. dissimilarity matrix and embedded in low dimensional Euclidean space. We define four categories or levels of similarity: ”excellent”, ”good”, ”fair”, and ”poor” with respect to a reference (or target) solution or a centroid measure in absence of a reference solution.

Graphical interpretation for the similarity scale is as follows:
Excellent Solution/s located inside the innermost circle
Good Solution/s located between the innermost and middle circle
Fair Solution/s located between the middle and outermost circle
Poor Solutions/s located outside the outermost circle



For more details on the technique:
1. Borg, I. and P. Groenen (2005). Modern Multidimensional Scaling, 2nd edition. New York, Springer
2. Razafindrakoto, H. N., Mai, P. M., Genton, M. G., Zhang, L., and Thingbaijam, K. K. (2015). Quantifying variability in earthquake rupture models using multidimensional scaling: application to the 2011 Tohoku earthquake. Geophysical Journal International, 202(1), 17-40.


source inversion validation database | version 1.0.0