21
$\begingroup$

I think Cross Validated is a more inclusive site than Data Science. The scope of Cross validated includes machine learning. I wonder what is the reason of this site proposals?

Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization.

and

Data Science Stack Exchange is a question and answer site for Data science professionals, Machine Learning specialists, and those interested in learning more about the field.

$\endgroup$
2

3 Answers 3

Reset to default
10
$\begingroup$

Yes, there's overlap, and with Stack Overflow too. If Cross Validated = Stats and Stack Overflow = engineering, and Data Science is stats + engineering, then that's fine.

I don't perceive Cross Validated as more inclusive. It's probably not a good place to ask about engineering model building at scale or data munging. Those might fit on Stack Overflow since it's pretty broad, but, do seem to have a better home here.

I agree it's still an open question whether there is enough distinct identity here to sustain an SE site. It's growing, but slowly. It's not obviously wrong though.

$\endgroup$
4
  • 2
    $\begingroup$ I say inclusive based on the site description on its tour page (Edited my question). I really don't agree with proliferation of site here at SE. I don't know about data munging but engineering model questions are being asked an getting answer at CrossValidated. $\endgroup$
    – Farid Cheraghi
    Sep 29, 2015 at 9:48
  • 2
    $\begingroup$ I see, Data Science has make a long way to get here and yes, it growing. But as a researcher I would prefer to deal with one Q&A site (CV or DataScience) for my scientific exploration.\ $\endgroup$
    – Farid Cheraghi
    Sep 29, 2015 at 9:50
  • 1
    $\begingroup$ @FaCh I also feel like there are a lot of sites. Too many and it dilutes discussions. This one is on the fence to me; it is a legit question. I think it's worth letting this site continue to evolve. In the meantime mods fairly aggressively kick out questions better on other SEs to try to keep the topic clear and focused $\endgroup$
    – Sean Owen
    Sep 29, 2015 at 10:16
  • 3
    $\begingroup$ I think DS should migrate questions to CV whenever the question is on-topic on CV. There is too much overlap at the moment, it's a waste of people time. $\endgroup$
    – Franck Dernoncourt
    Mar 7, 2017 at 21:40
5
$\begingroup$

I agree with @SeanOwen (+1), but I would use even stronger words, because the difference is exactly the opposite from what you are suggesting. Clearly, it is data science that is a much larger domain, comprised of many disciplines and related technologies, areas and aspects, where statistics (the focus of Cross Validated) is just a part, even though a critically important one.

Therefore, while Data Science SE formally is "inclusive", as you put it, separation of content by focus, as Sean noted, leads, in my opinion, to existing best practices: questions, focusing on programming, best fit StackOverflow; questions, focusing on statistics, best fit Cross Validated; questions, focusing on data science aspects, with exceptions of the two above-mentioned, best fit Data Science SE (i.e., data manipulation, model building, machine learning, big data tools, data workflows, data science infrastructure and, I would say, even the relevant career-related topics).

$\endgroup$
1
  • 5
    $\begingroup$ Machine learning is squarely on-topic on stats.SE. Many aspects of data manipulation are also on topic (e.g. different strategies for encoding categorical data; feature selection and extraction methods). I don't know what, precisely, you mean by "model-building" but I expect that most of it is also on-topic on stats.SE. $\endgroup$
    – user45817
    Aug 13, 2018 at 14:45
0
$\begingroup$

I used to have a similar doubt, when I was starting to use both the sites.

But, after experiencing enough of them, I had self-answered my question on the Meta SE site. And your question is eerily similar to mine. (But, it is a very popular question for starters of both the communities.)

Data Science SE as @AleksandrBlekh and @SeanOwen has nicely put, has a lot of content overlapping with CrossValidated.

But, a significant chunk of questions and posts here, would not be on-topic on CV and SO.

For example: my question on Columnar Databases in the Data science domain, would be off-topic elsewhere but here; as it specifically asks for this domain, and is also very important and prominent concept.

So, the posts about engineering side of Data Science are the crux of this community, and those are the ones which make this site unique.

$\endgroup$
2
  • 2
    $\begingroup$ Your sample question is very on-topic on www.stackoverflow.com website. There are a lot of overlap between stackoverflow and cross-validated with Data Science. This is the reason of slow growing of this website. $\endgroup$
    – Farid Cheraghi
    Oct 1, 2015 at 10:26
  • 1
    $\begingroup$ Yes, I totally agree with the slow growth. But, I definitely think it has been quite active in recent times. This discussion is also worth looking at: meta.stats.stackexchange.com/questions/2253/… $\endgroup$
    – Dawny33
    Oct 1, 2015 at 10:43

You must log in to answer this question.

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