I am considering a specific topic in research paper within data mining field. I just found that my idea is being implemented and published in arXiv in the last month. So far, the published paper matches the flow that I was planning to follow but the author has considered the study with Twitter data. My question is: If I want to do the same experiments but with a different data set like facebook data set will it be considered a new contribution in that field ?

Ps: Row data crawled from Twitter API is way different from row data crawled from Facebook API, and the also different in structure and body.

1 Answer 1


In fact, the main contributions in your field are "algorithms", however, data sets are just a bunch of inputs for testing those algorithms and assessing their functionality or efficiency. If the outcomes of your new tests with data set A are exactly the same as dataset B's ones, you might not claim it as a new contribution. But if the results acquired from dataset B lead to some different interpretations about your algorithm's computational capabilities or something, you can consider it as a new contribution.

  • you mean by algorithms - ''machine learning'' ?
    – Krebto
    Commented Mar 6, 2017 at 19:22
  • 1
    @Krebto: Yes. I mean algorithms are often contributions of machine learning and most of the CS-based research fields.
    – user41207
    Commented Mar 6, 2017 at 19:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .