Classifiation SubProcess for Feature Selection

Muhammed_Fatih_

Hello together, 

do you have a recommendation with regard to the question of which classification model sould be used within Feature Selection (e.g. Optimize Selection or Backward Elimination) to be able to efficiently select attributes or rather dimensions based on a high-dimensional TF-IDF matrix? 

Thank you in advance for your answers! 

Best regards, 

Fatih

varunm1

Hello @Muhammed_Fatih_

A general approach is to use a model that is used as base learner in your process. For example, if you are trying to make an SVM based prediction model then it is better to have SVM inside feature selection subprocess as well.