So my question is how do I declare a new attribute for weka when all my data is a collection of texts.
My goal is to create new attributes, so that I can use a decision tree algorithm. For example, I want an attribute that Filters for all these Keywords (Winner, Free, Subscription)
If you are thinking of generating new attributes from the textual data in your
STRINGattribute, then you can use the StringToWordVector filter.For general info on handling textual data, have a look at the Text categorization with Weka wiki article.