Proposal: Big Data

How is "Big Data" distinct from the stats stack exchange? It seems to me like questions of theory could be asked on https://stats.stackexchange.com/ and technical questions could be asked on stackoverflow.

  • could you say more on 'questions of theory could be asked on stats.stackexchange.com'? I thought big data was too much data for db's so I'm not sure how big data questions would be statistics questions. – Duncan Jan 14 '13 at 11:56
  • @Duncan stats.stackexchange is for both statistics AND machine learning including building models incrementally, processing data sequentially etc. – steffen Jan 26 '13 at 6:43
  • 1
    A lot of statisticians come from academic backgrounds where they use R or similar tools to analyze small(ish) data sets. While these people are knowledgeable about stats, they probably have no idea how to efficiently scale their calculations to datasets that can only fit on multiple computers. It is true that Big Data overlaps with many other fields (low-/high-level programming, IT, stats, ML...), but the Big Data experts of these specific fields usually possess knowledge that is somewhat (or completely) different from the knowledge possessed by the non-Big Data experts of those same fields. – Felix GV Feb 5 '13 at 0:29

Perhaps creating this "BigData" we could put together people who have this commmon need of knowing how to deal with archiving BigData and then extracting stats from it.

Probably questions that are fitted in this site could also be asked in others, but there's a great probability that in this site we would aggregate more experts in the subject...

  • Speaking as the devil's advocate: This is exactly the same argument the defenders of the machine-learning-SE have stated. The site-proposal has failed, because not enough contributors show up, most decided to stay in stackoverflow and stats.SE. So if ML.SE has failed and big data is somehow a subdiscipline of ML ... you can do the math ;). – steffen Jan 26 '13 at 6:49
  • 4
    Saying Big Data is a subdiscipline of ML probably reflects a misunderstanding of what Big Data is :) – Felix GV Feb 5 '13 at 0:20

You must log in to answer this question.

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