Doc2vec is an extension of word2vec, which creates vector representations of documents.
One can use this representations as input to some classifier/regression(Logistic Regression, XGboost, LightGBM ...).
What about using the similarity as a prediction. Let's say you define some typical examples of classes through clustering or by choice and than you similarity to these examples as a prediction.
Is that a decent plan? Is it something that is used with doc2vec/word2vec.
