====CDMC 2017: Call for Participation====
The 8th International Cybersecurity Data Mining Competition (CDMC 2017)
http://www.csmining.org/cdmc2017/
The CDMC 2017 is a challenging, multi-month research and practice competition, focusing on the application of knowledge discovery techniques to solve advanced, real-world problems. The competition is open to teams worldwide. Contestants are invited to participate in solving a set of problems in fields of cybersecurity, text mining, image processing, etc.
The competition is associated with the 10th International Workshop on Data Mining and Cybersecurity (DMC 2017), Which is an associated event to the 24th International Conference on Neural Information Processing (ICONIP 2017), Guangzhou, China. The competition is open to anyone who wishes to participate.
Competition starts: 1st July 2017
PRIZES & AWARD
We have set cash prizes for the competition. The top ranking team of all data mining tasks will be eligible to win a cash prize of NZD $3000. Additional prizes may be available as travel grants for deserving participants to help them attend the ICONIP2017 conference and the DMC2017 workshop. The award ceremony of the competition will be held at the grand banquet of ICONIP, where the winner of the CDMC will be awarded in front of all the audience and honorable guests of the conference.
IMPORTANT DATES
Competition starts: 1st July 2017
Competition closes: 15th October 2017
Winner Announcement: 5th November 2017
Award Ceremony: 17th November 2017 (ICONIP Banquet)
Conference: 14th-18th November 2017
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 CDMC 2017 website to get the competition tasks datasets. The method of result submission is via the form on the Upload page. To be ranked, submissions must include only results on the test portion of data sets. Multiple submissions are allowed.
SUBMISSION:
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 (October 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.
--
Ruibin Zhang
Department of Computing and Information Technology
Faculty of Creative Industries and Business, Unitec, New Zealand
Email: pzhang(at)unitec(dot)ac(dot)nz
|