4. 2016.10.21 Analysing Importance of Variables in Models

using R-lang

Paper: http://www3.dsi.uminho.pt/pcortez/2010-rminer.pdf

• Library: rminer

  library(rminer)
  fit = fit(y ~ x1 + x2 + x3 + x4 + x5 + x7 + x8, 
              sampled_train_data, model = 'svm')
  im = Importance(fit,sampled_train_data,method='sensv')
  

• In fit() func, we can train many different kinds of models, by editing the model= argument

• By specifying the method arg, we can select many different kinds of methods, such as 1D-SA, DSA, MSA, etc