I have a dataset (excel file) that contains a customer reviews. I have to classify these review into seven predifined categories according to their content. But some sentence may belongs to more than one category.
How can I handle this issue in KNIME ? Is there any example workflow ? Which nodes should I use ?
I cant get clearance in my mind. Your helps are appreciated.
I’ve never done text classification, but there are plenty of examples on the KNIME Hub: https://hub.knime.com/search?q=text%20classification
I guess the hub should be very helpful. In general you probably want to convert your text into numbers (tokenization, embedding) and if you go with keras then I suppose a final layer with 7 outputs and a sigmoid function)
This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.