I want to analyse user reviews for certain products as part of a research project without having to learn analytics from scratch, as my requirement is temporary. I need to do the following:

  1. The user reviews are formatted in separate columns per user and indexed over multiple pages (almost 2000 reviews per product) - so first extract all the text intelligibly.
  2. Analyse frequency of certain words and phrases(once per user), the overall connotation of the comment, most frequent words of all the reviews combined, etc.


  • Where do I begin doing this?
  • How should a beginner approach this?
  • How long is it going to take?
  • Could there be an easy to use software or a bot where I can just give in my requirements and would just give me the results?

1 Answer 1


Welcome to DS.SE, @shubham.

Since the question is a bit general, I can just provide you some link to start with. Once you have more specific requirements, the community can help further.

An easy-to-use tool that I knew for sentiment analysis was IBM Alchemy which seems retired now in favor of IBM Watson NLU.

Despite IBM, other analytical platforms such also provide sentiment analysis APIs/tools. For example, see Google AI & Machine Learning Products and Deep AI. This article on Medium reviews some this tools in brief.


Not the answer you're looking for? Browse other questions tagged or ask your own question.