R survivalsvm package

Question

[haiying.kong]

Dear All,

I am trying to use the R package: survivalsvm. I can get it running without error, survivalsvm and predict.survival, but if I look at the predicted values from the output of predict.survivalsvm, the values are not easy to interpret. They did not look like expected relapse time, or probability for relapsing. Can anyone please help me?

[h.mon]

I believe you will hardly get any help, unless you explain your data and show how you run the analysis.

[haiying.kong]

Not really. It is a general question about the software tool.

For instance, if I run the example code on the manual of the software, the predicted values are:

pred.survsvm.reg$predicted [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [1,] 13.88731 14.94699 11.12112 15.60472 10.69572 13.52772 10.89338 10.51362 [,9] [,10] [,11] [,12] [,13] [,14] [,15] [,16] [1,] 15.1629 11.15693 10.58539 9.820887 14.3924 9.964974 14.94568 10.24687 [,17] [,18] [,19] [,20] [,21] [,22] [,23] [,24] [1,] 10.16142 10.89258 9.418691 11.57923 11.11941 11.24054 10.17788 10.31566 [,25] [,26] [,27] [,28] [,29] [,30] [,31] [,32] [1,] 14.60489 10.65962 12.97571 11.03749 10.24398 10.29754 14.29548 9.434277 [,33] [,34] [,35] [,36] [,37] [,38] [,39] [,40] [1,] 9.25484 9.238892 9.121091 9.392231 10.04443 9.571002 14.50668 15.25616 [,41] [,42] [1,] 11.59102 14.39756

How to interpret this?

This is really a general question.

[antass]

+1 I have the same question. A very promising package with a lacking manual / vignette which makes it unusable :/