I have a dataset consisting of questions from customers. I am curious of the n most frequent asked questions, regardless of the variation the questions might appear in.

Is there NLP methods for finding n most frequent asked questions?

  • $\begingroup$ How are you measuring "closeness" for questions? If you are looking for a general answer, you probably can use a pre-trained NLP model like bert - towardsdatascience.com/…. If you want something specific, can you provide more details? $\endgroup$ Commented May 14, 2022 at 1:39
  • $\begingroup$ How do you define variation of the same question and different question? $\endgroup$
    – emely_pi
    Commented May 23, 2022 at 13:25
  • $\begingroup$ How I measure closeness, variation or distance is part of the question. I know its common to use cosine to measure similarities in semantic spaces. But I'm curious if there were some methods I've overlooked for this problem. $\endgroup$ Commented May 24, 2022 at 14:30


You must log in to answer this question.