I am working on a prediction model where I have several input variables. I would like to eliminate variable columns which decrease the model efficiency. I went through your post https://tech.knime.org/forum/knime-general/variable-importance-in-prediction-classification-or-regression-molels where a sample workflow has been attached to get the variable importance. I would like to understand it in a better way. Does this model indicate that if I remove the columns having least accuracy from my original prediction model, the overall accuracy would increase? My target is to eliminate the columns which decrease the overall accuracy and present the model with important input variables.