< September 2017 >
Su Mo Tu We Th Fr Sa
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30

Steps to participation:

Anybody who complies with the rules of the competition (CDMC 2017) is welcome to participate. All participants must register their identities on the CDMC2017 website to get the competition tasks datasets. The method of result submission is via the form on the Submission page. To be ranked, submissions must include only results on test portion of data sets. Multiple submissions are allowed. The steps are summerized as follows:

1. Registration → 2. Login → 3. Data & Task → 4. Submission

Submissions:

For each task results submission, you are required to submit only the class label of the test data in a plain TXT format file. Note that the class label is required to keep the same definition as the training dataset.  In addition to submission, you are encouraged to submit a 2-8 page short paper to the DMC2017 workshop.

Evaluation:

The results of classification will be represented in a confusion matrix composed of tp (true positive), fn (false negative), tn (true negative), and fp (false positive), respectively. We define the precision and recall as: Precision =tp/(tp+fp), Recall=tp/(tp+fn). Then, the results are evaluated by calculating F-measure as F=2*(Precision*recall)/(Precision+Recall).

Ranking:

For each team that participates in CDMC2017, only their last valid entry will count towards determining the winner. Valid entries must include results on all three competition tasks. Prizes will be attributed only to entries performing better than the baseline method. The results of the baseline method will be provided on the site before the competition end date (September, 2017).

Reproducibility:

Only top ranking participants  will be required to fill out a fact sheet about their methods.  If successful, they will have to make a  contribution towards publishing their paper in the proceedings and would need to give permission for their methods to be implemented if required.

Prizes:

We have set a cash prize for the competition. Only results on all 3 tasks test datasets will count towards the winning prize. The top ranking team of all 3 data mining tasks will be eligible to win a cash prize of $3000NZD.