Is it possible to derive rules from a data set that is classified?

I’m new to Knime and facing an issue with rules.

The datasets available in Kaggle for multiclass classification have the class assigned, however, the rules are not easily identifiable or derivable just by observation.

Dataset snapshot of training data-

The last column- Segmentation is the class that needs to be predicted.

Could you let me know what approach I can use to proceed with this scenario?

Thanks in advance.


@snethala welome to the KNIME forum. If you want to solve multiclass problems I can offer this example. Please also check the additional links beneath the workflow picture under “External resources”.

You can read the links in this collection marked multiclass:

You might also want to read about machine learning in general. Maybe you can give us an idea about your level of experience.

1 Like

If you’re not sure where to start you might try the AutoML approach. It fits a number of models and “picks” the best one. AutoML is included in KAP 5.2.

1 Like