Approach-
-
We created dependency tree using NLTK to understand the text data.
-
We used SpaCy to extract Subject from the text and based on the text we found their related Adjective, Preposition etc by iterating over the tree. We stored them in a dictionary and this dictionary is nothing but the opinion or reviews.