Feature selection and Model Tuning in R



Hello everyone,

Th following modifications are done to prepare the data ;

  1. combine the train and test set data
  2. missing value treatment
  3. feature scaling i.e combine levels within categorical variable,make new variables for numerical predictors
  4. feature extraction
  5. tune the model for lm, glm, rpart, rf

But still my RMSE is 1150 on the test data set. Anyone can suggest me what I do next ?

Thanks !!!


There are various things which you can try:

Also you can do some visualisation before in order to get some insights, which can be further useful in creating new features.

Further, you can read different winner’s approach from here,

Hope this helps.