Table contains no nominal attribute for classification.

[Sabina]

Hello,
I am new to the KNIME Analytics Platform, so I am still learning and troubleshooting what are probably basic problems. My data set is in Excel, so I used the Excel Learned Node, and (after partitioning it in a training and test set) I was going to put in the Decision Tree Learner node, but I got the error that says “Table contains no nominal attribute for classification”. I’ve read that this error occurs when there is a difference between the domain of the training data and the domain of the test data. Has anyone run into this issue before and can advise what the solution may be? Thank you in advance!

[JanDuo]

Hi @Sabina
Welcome to the Knime Community!

The Decision Tree Learner node expects as target a nominal attribute. Nominal means that it is not numerical. So a column with e.g. values “red”, “green”, “blue”. Or “A”, “B”, “C”.
When your Excelsheet contains only numerical columns you will get this error.

When you want to predict a numerical value, you might want to use the Linear Regression Learner node instead.

[Sajjadmosaheb]

hi @Sabina
As @JanDuo said and explained what is nominal
Also, if your data is ranked, you can use Number To String node and change the target column to string .