Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

I thought it would be interesting to plot the 'experience level,' as determined by medals or number of posts, vs calendar time for questions tagged with "R" (posted to SO ). Assuming I teach myself some of the R-packages for general data mining, is there an accessible dbase within SO that could be used for such a purpose? (Not that I would turn down a fully functional solution to this problem :-) )

share|improve this question

migrated from Nov 23 '12 at 14:47

This question came from our site for professional and enthusiast programmers.

Maybe this? – EDi Nov 23 '12 at 13:33
I have no objection to moving the question -- I figured it was rather "meta" . – Carl Witthoft Nov 23 '12 at 13:59
or – Ben Bolker Nov 23 '12 at 14:02
Thanks to all commenters and answerers. Off I go to scrape data. – Carl Witthoft Nov 23 '12 at 18:41
up vote 4 down vote accepted

The best place to find the data would be in Predict Closed Questions on Stack Overflow. This is a preprocessed dataset from Stack Overflow to predict if a question will be closed or not.

The next resource is the Stack Overflow Creative Commons dump.

And to get an insight into some fun stuff you could do with the data, check out the paper Discovering Value from Community Activity on Focused Question Answering Sites: A Case Study of Stack Overflow (published 2012-08-12) from a bunch of heavy hitters.

share|improve this answer

Stack Overflow has been publishing "data dumps" for quite some time, as well as offering an API for interactive exploration. Note that the full data dump is really large. The download is IIRC around 5 GB of 7-Zip compressed XML data; in order to fully decompress it you thus need a lot of disk space.

The Kaggle data set, Predict Closed Questions on Stack Overflow, might therefore be an easier start.

Note however the rather obvious privacy implications. The publicly available data of Stack Overflow is to some extend anonymized, but probably not to the extend that makes your efforts moot though.

Nevertheless, do not undererstimate the difficulties you will encounter.

share|improve this answer
Did you just copy paste my response ? and add privacy concerns :) – steve Nov 23 '12 at 14:24
@steve I meant to primarily answer with the privacy implications, then noticed that I maybe should answer the question first before adding them. But actually I have downloaded and used the data dump. Which is why I started with that link and know the size from last year. ;-) But I didn't find anything interesting for me back then; I was looking into frequent tag combinations, but that was all trivial stuff, and the 5 tag limit bias showed heavily. I had seen the kaggle, too, but not participated. – Anony-Mousse Nov 23 '12 at 17:02

You must log in to answer this question.

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