The R Journal: accepted article

This article will be copy edited and may be changed before publication.

Support Vector Machines for Survival Analysis with R PDF download
Césaire J. K. Fouodo, Inke R. König, Claus Weihs, Andreas Ziegler and Marvin N. Wright

Abstract This article introduces the R package survivalsvm, implementing support vector machines for survival analysis. Three approaches are available in the package: The regression approach takes censoring into account when formulating the inequality constraints of the support vector problem. In the ranking approach, the inequality constraints set the objective to maximize the concordance index for comparable pairs of observations. The hybrid approach combines the regression and ranking constraints in a single model. We describe survival support vector machines and their implementation, provide examples and compare the prediction performance with the Cox proportional hazards model, random survival forests and gradient boosting using several real datasets. On these datasets, survival support vector machines perform on par with the reference methods.

Received: 2017-10-01; online 2018-05-16, supplementary material, (11.2 Kb)
CRAN packages: survivalsvm, kernlab, pracma, quadprog, Matrix, randomForestSRC, mboost, mlr, ggplot2, tikzDevice
CRAN Task Views implied by cited CRAN packages: MachineLearning, Multivariate, NumericalMathematics, Optimization, Survival, Cluster, DifferentialEquations, Econometrics, Graphics, HighPerformanceComputing, NaturalLanguageProcessing, Phylogenetics, ReproducibleResearch


CC BY 4.0
This article and supplementary materials are licensed under a Creative Commons Attribution 4.0 International license.

@article{RJ-2018-005,
  author = {Césaire J. K. Fouodo and Inke R. König and Claus Weihs and
          Andreas Ziegler and Marvin N. Wright},
  title = {{Support Vector Machines for Survival Analysis with R}},
  year = {2018},
  journal = {{The R Journal}},
  url = {https://journal.r-project.org/archive/2018/RJ-2018-005/index.html}
}