I am trying to do logistic regression but had gotten this warning msg any one know how i can solve it
WARN Logistic Regression Learner 9:5 Columns “country”, “reservation_status_date” have too many different values - will be ignored during training (enforce inclusion by using a domain calculator node before)
Thank you
Have you tried another classifier which is less sensitive to this problem?
Hi
No was new to KNIME so not really sure what should i do
Without knowing anything about your data and what you’re trying to do, its difficult to make specific suggestions. If you have lots of variables (forget values for the moment) you run a high risk of overfitting. This isn’t a KNIME issue per se. How much do you know about classification algorithms?
The country column should be a categorical variable in logistic regression and would require you to fit a dummy a variable (k-1) i.e. for each country excluding one country as the first or last category . You would need to think about how you are going to use the reservation_status_date, i.e., use it to calculate a new variable, it has no value on it own .