Welcome to the Simplilearn Community

Want to join the rest of our members? Sign up right away!

Sign Up

Machine Learning Advanced Certification | Rishabh Kaushal | Jan 23 - Feb 27 | 2021

Hello Team, I am planning to work on the Amazon-Netflix project ? I realized a couple of you might be interested in working on that project.

Feel free to IM me in next live session if you think it would help to work as a group and share our learnings to each other or send me an email at bhargav.tri15@gmail.com

Neel Amin

Dear friends,

I was working on Amazon-Neflix Project was not sure on the following:

  • Define the top 5 movies with the least audience (Does this mean with the list no. of users who have rated Eg: Movie1 has 3 counts of Ratings; Move2 has 1 count of Ratings and Movie3 has 4 count of ratings then Movie2 has the least audience?
  • Make predictions on the test data --> Where do we pass the test data ??
Last edited:
Hello Friends,

I have a question. If in a given dataset, where columns names are masked(we don't know what is what), I would like to know if it is appropriate to remove the column(s) having all values as unique?
I am asking this based on the logic that IDs and similar unique features do not contribute to model, so it is appropriate? Can there be ever a case in regression problem where the target variables have all unique values, but we don't know in case the column names are masked in dataset on what column is related to what. Please let me know your thoughts