I extracted the most important features from the text dataset then I did different steps related to preprocessing, query expansion, TFIDF, and computing the similarity using cosine. My dataset became a DataFrame containing 14 columns, those columns are the features of the dataset. My question is :