AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

Kaggle
 
 

Recently came across Kaggle, which contains different AI related competitions. They aren’t really ‘chatbot’ oriented, except for the digit recognizer competition perhaps, but it should be an interesting challenge for the AI developers out there.

 

 
  [ # 1 ]

Wow! some of those prizes are worth winning! The smallest one I saw (apart from kudos) was $10,000

 

 
  [ # 2 ]

My thoughts exactly. The competition is probably pretty impressive as well.

 

 
  [ # 3 ]

I entered my first kaggle competition recently. I plan to enter many more. This first one had a restriction of solo-teams, but there are lots of competitions that allow for teaming up. If anybody’s interested in this, drop me a note.

 

 
  [ # 4 ]

Good luck! Is there a way we can vote for you to boost your ranking?

 

 
  [ # 5 ]
Steve Worswick - Sep 6, 2012:

Good luck! Is there a way we can vote for you to boost your ranking?

Thanks.
No, the ranking is just temporarily.  It all has to do with how Kaggle is set up. All competitions have the same format: you get a (set of) training file(s) and a test file In this case there was only 1 training file, which contained statements that they had already labeled as either insulting or not. The test file doesn’t have this labeling, the idea is that you generate a result set, based on the test data and send this to kaggle. That’s what they use to show the ranking.
At the end of the competition though, everybody needs to generate another result set, based on a new test file that will be given at that time. This is to prevent people from manually labeling things and for catching ‘overfitting’ on the test data.
So the whole order can still change. I also haven’t included a spell checker yet (I’m working on that right now, don’t know if that will be ready in time for this competition though).

The interesting thing about these challenges is that it’s a narrowly defined field + you get lots of training data.

 

 
  [ # 6 ]

I read one contest was to analyze 6GB of data.
Seemed like quite a heavy workload for a contest?

But I signed up anyway.

 

 
  [ # 7 ]

Well, it’s ‘big data’.  (great to test the limits of my app also)
Which contest did you enter?  I already did 2, now on my 3th. Learned an enormous amount of new stuff (currently doing the random forest/ decision tree stuff).

 

 
  login or register to react