2

Possible Duplicate:
Locating freely available data samples
Sites for predictive modeling competitions

Are there sites or sources of "datasets" (either artificially created or taken from actual experiments/sources) someone can use to test their own abilities in detecting clusters, patterns, and to formulate/validate hypothesis?

Possibly with answers, that could be consulted, later to determine what you got and what the expected results/techniques should have been?

Community
  • 1
p.marino
  • 83
  • 5
  • 2
    You may be interested in this question: [locating-freely-available-data-samples](http://stats.stackexchange.com/questions/7/), or this: [tiny-real-datasets-for-giving-examples-in-class](http://stats.stackexchange.com/questions/5937/), or just search under the [tag:dataset] tag. – gung - Reinstate Monica Oct 23 '12 at 15:05
  • Thanks to both. I have zero skills in the field (but I'd like to get better thanks to books and coursera-style courses) so I am mostly interested in practical "exercises", than in competitions. So the first comment is closer to what I am looking for, but thanks (and upvotes) to both of you ;) – p.marino Oct 23 '12 at 16:47

2 Answers2

2

Kaggle hosts a variety of data mining competitions. In addition to their active competitions, they have archives of past competitions. Not only can you compare your results with the correct answers, but you can also compare your results to the winners of those past contests. I'm not sure if the scope is appropriate for what you are after though...these probably require a bit more effort than a typical "quiz" would.

Michael McGowan
  • 4,561
  • 3
  • 31
  • 46
2

There is a free introductory textbook at openintro that also has sample datasets and a library of homework/quiz questions. That would be a place to start. I don't know if it will get into as advanced topics as what you want.

Greg Snow
  • 46,563
  • 2
  • 90
  • 159