The problem is predicting hits on different stories published on a website. I am aware that for this kind of time series forecasting, facebook prophet is a popular library. However, it seems in prophet, each model is trained on one time series which is used to predict the same time series in the future.

But if I have a site publishing different stories, and I need to predict the views on the stories, clearly it is infeasible to train one model for each story. Also, as training data, I have the history of only past stories on how their popularity evolved after publication.

So is prophet suitable for this scenario? Or any other popular time series library?

  • $\begingroup$ I think prophet is the wrong tool for this work. We would like to reformulate this as a regression problem that entails estimating the expected number of hits during a particular time interval (e.g. first 2 weeks). The features here being the story characteristics (length, subjects, themes, photos, etc.). $\endgroup$
    – usεr11852
    Apr 11, 2021 at 0:55
  • 1
    $\begingroup$ Can you add some more information, like possible covariables, number of stories (so far), and what kind of prediction you need, what is the purpose? $\endgroup$ Apr 11, 2021 at 1:02


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.