random forest using ID as a predictor

General
1

Hello!
I am running a random forest model , my data includes an ID which is being used as a predictor. how do I keep it in the dataset but not use it as a predictor?
Note: This ID is appearing as an important factor yet it should not even be relevant in the prediction.
Please help.

snippet of the random forest
classifier = randomForest(dpt3~., data=train,
importance = TRUE, ntree = 1000,mtry = 4,
replace=TRUE, random_state = 0)

when i check variable importance here is what i get, it is the top.
randomForest::importance(classifier)
1 2 3 MeanDecreaseAccuracy MeanDecreaseGini
public_id 60.485515 97.234941 79.874132 149.191509 2258.26709

2

Generally speaking in model fitting in R with functions that use formulas you can use minus symbol to exclude a variable from being a predictor

classifier = randomForest(dpt3 ~ . -ID, data=train,
1 Like
4

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.