Data Mining Cup 2011

I would like to tell you about the data analysis competition that is about to start. Full information about him can be found on the official website . Only students can participate in the competition, so if you are a student and you are interested in data analysis, then you can register (email can be regular, not university, if there is none) and participate.

Next you will find the schedule of the competition, the basic rules of this competition. The same link to the conditions of previous years (for training and those who are no longer students).



DMC 2011 Competition Schedule

Event	Location	date
Registration for the DMC 2011 competition begins	online	March 01, 2011 to May 31, 2011
Start of the DMC 2011 competition	online	April 15, 2011
Last date for sending DMC 2011 results	online	May 31, 2011
Award winners DMC 2011	Pentahotel Leipzig, Germany	June 21, 2011 to June 22, 2011

Fundamental rules:

1. Only students can participate
2. Maximum 2 teams from the university.
3. Registers the head of the team, he is responsible for the whole team, and will receive an award in case of winning.
4. There are no restrictions on team size.
5. Prizes:
   • 1st place - 2500 EUR
   • 2nd place - 1500 EUR
   • 3rd place - 1000 EUR

The remaining rules are less significant (or relate to the above dates) can be found by reference .



There are no conditions, as you understood from the timetable, so for preparation, or for those who are no longer a student, but are interested, I suggest looking at the conditions of previous years .

For example, I worked with DMC 2006 data. This data was for iPod sales on eBay in the German sector. And it was necessary to build a model that would determine whether the bid was sold above or below the average price in the category. The juice itself was in the analysis of textual variables TITLE and SUBTITLE - in which various information was stored, and the text itself was typed, without a clear format.

I dealt with this data for a while - and the best that I received with the help of Gradient Boosting was # 8 place. If interested, I can write a more detailed article about what I did and how.