The attribute abandon was already present in the example set

I am working on building a classification model for classifying the text as positive, negative or neutral using  Linear SVM. i have trained it using a pre labelled dataset. But on testing the model shows an error as "The attribute abandon was already present in the example set" in "Process documents from data" operator. have tried all options for keeping the "keep text" ON/OFF but nothing worked. How to resolve it?