Matches a given question to the most similar questions in a question pool.
Create a questions.csv
in the form of question, answer, [choice1], [choice2], [choice3], with the ones in [] being optional.
Then when you run main()
after re-loading the csv, it will prompt you for a question, fill it in, press enter, and it'll give you the most similar questions in your questions.csv
.
Uses sklearn's TfidfVectorizer under feature extraction for text, with nltk.corpus' stopwords to preprocess text.