Dear all, I would like to use a document classifier and I found that OpenNLP provides one, but there is no information in documentation about the features used in classifying the document. Does it depends on Term presence, Term frequency, or TF-IDF. Actually it seems that the classifier depends on the bag-of-words instead of terms since I do not need to transfer the document to vector at all. I might misunderstand or miss something. So, please can you clarify the features on which the document categorizer depends? Best,Amal
