{ "id": 8331, "title": "Date use for teaching after competition concludes", "url": "/competitions/decoding-the-human-brain/discussion/8331", "author_name": "", "post_date": "2014-05-30T14:49:39.703Z", "votes": null, "comment_count": 7, "views": 2802, "content": "

Hi! After the competition has concluded, am I allowed to use the data from this competition as an example dataset for teaching? I ask because this would technically involve students discussing the data, and potentially me giving them some starter code, that would be unavailable to others who want to attempt the competition. Since I'm strictly asking with the idea of doing this *after* the competition has completed, I don't think this constitutes an unfair advantage to anyone. What do you think?

", "messages": [ { "id": "47340", "postDate": "05/30/2014 14:49:39", "content": "

Hi! After the competition has concluded, am I allowed to use the data from this competition as an example dataset for teaching? I ask because this would technically involve students discussing the data, and potentially me giving them some starter code, that would be unavailable to others who want to attempt the competition. Since I'm strictly asking with the idea of doing this *after* the competition has completed, I don't think this constitutes an unfair advantage to anyone. What do you think?

", "rawMarkdown": "", "votes": null }, { "id": "47341", "postDate": "05/30/2014 15:00:19", "content": "

Hi,

\n

As stated in the rules page, the dataset is distributed under a Creative Commons license, so you are free to use it, as long as you provide attribution. Moreover, we designed the distribution license to encourage the re-use of the dataset after the competition. Teaching is for sure an important context in which the dataset can be of help. We would be glad if you do that.

\n

As a side note, we also provide example source code (Python and Matlab) to load, pre-process, analyse and visualise the dataset. The code is distributed under an Free / Open Source license (BSD, 3 clauses). It may be of help for teaching as well.

", "rawMarkdown": "", "votes": null }, { "id": "47342", "postDate": "05/30/2014 15:06:56", "content": "

Thanks for the quick response! The issue I was worried about was if the students submit their solutions to the Kaggle website for scoring, after the competition had concluded. Does that make sense?

", "rawMarkdown": "", "votes": null }, { "id": "47349", "postDate": "05/30/2014 17:12:08", "content": "

Hi cMadan, thanks for checking.

\n

After this competition deadline passes (27 July 2014 11:59 PM UTC) the leaderboard becomes frozen and the winners will be determined from that point in time according to those official Rules.  So if you begin discussing the problem in class after 28 July (and/or providing starter code) to your students then, it won't interfere with the fairness of the competition. [If your students discuss code or methods offline before 28 July, this would be breaking the rules of the live competition and they would be disqualified.]

\n

After a competition closes, Kaggle actually supports ongoing learning in precisely the way you're hoping. Post-27 July, we leave the scoring mechanism enabled, so that anyone on Kaggle can continue to practice with new submissions and see what their score "would have been".  These submissions don't get a placement on the leaderboard, but do give individual feedback scores on their My Submissions page.

\n

I hope this helps!

", "rawMarkdown": "", "votes": null }, { "id": "47351", "postDate": "05/30/2014 17:34:41", "content": "

Hi Ramzi, thank you for your detailed answer! My plan is to use the example after the competition is completed, so this is perfect! Thanks again!

", "rawMarkdown": "", "votes": null }, { "id": "63475", "postDate": "02/03/2015 10:07:25", "content": "

Is this dataset used for any other published paper in a journal or conference? Please, if you know, kindly refer me to it. 

", "rawMarkdown": "", "votes": null }, { "id": "63478", "postDate": "02/03/2015 10:42:52", "content": "

Hi Fatima,

\n

Dan Wakeman and Rik Henson, the authors of the original dataset from which the one of the competition was extracted, recently published this paper: http://www.nature.com/articles/sdata20151

\n

Additionally, last December we prepared a short article for the Machine Learning In NeuroImaging (MLINI) workshop of NIPS (http://nips.cc/Conferences/2014/Program/event.php?ID=4307). There, a portion of the competition dataset was used to test a proposed algorithm. The paper was accepted and it is expected to be published in the workshop's proceedings in 2015.Best,Emanuele

", "rawMarkdown": "", "votes": null }, { "id": "63555", "postDate": "02/04/2015 05:54:13", "content": "

Thank you for the quick reply. It helped me alot. 

", "rawMarkdown": "", "votes": null } ], "comments": [ { "id": 47341, "author_name": "emanuele", "author_url": "", "post_date": "05/30/2014 15:00:19", "content": "

Hi,

\n

As stated in the rules page, the dataset is distributed under a Creative Commons license, so you are free to use it, as long as you provide attribution. Moreover, we designed the distribution license to encourage the re-use of the dataset after the competition. Teaching is for sure an important context in which the dataset can be of help. We would be glad if you do that.

\n

As a side note, we also provide example source code (Python and Matlab) to load, pre-process, analyse and visualise the dataset. The code is distributed under an Free / Open Source license (BSD, 3 clauses). It may be of help for teaching as well.

", "votes": null, "replies": [] }, { "id": 47342, "author_name": "cmadan", "author_url": "", "post_date": "05/30/2014 15:06:56", "content": "

Thanks for the quick response! The issue I was worried about was if the students submit their solutions to the Kaggle website for scoring, after the competition had concluded. Does that make sense?

", "votes": null, "replies": [] }, { "id": 47349, "author_name": "rjrest", "author_url": "", "post_date": "05/30/2014 17:12:08", "content": "

Hi cMadan, thanks for checking.

\n

After this competition deadline passes (27 July 2014 11:59 PM UTC) the leaderboard becomes frozen and the winners will be determined from that point in time according to those official Rules.  So if you begin discussing the problem in class after 28 July (and/or providing starter code) to your students then, it won't interfere with the fairness of the competition. [If your students discuss code or methods offline before 28 July, this would be breaking the rules of the live competition and they would be disqualified.]

\n

After a competition closes, Kaggle actually supports ongoing learning in precisely the way you're hoping. Post-27 July, we leave the scoring mechanism enabled, so that anyone on Kaggle can continue to practice with new submissions and see what their score "would have been".  These submissions don't get a placement on the leaderboard, but do give individual feedback scores on their My Submissions page.

\n

I hope this helps!

", "votes": null, "replies": [] }, { "id": 47351, "author_name": "cmadan", "author_url": "", "post_date": "05/30/2014 17:34:41", "content": "

Hi Ramzi, thank you for your detailed answer! My plan is to use the example after the competition is completed, so this is perfect! Thanks again!

", "votes": null, "replies": [] }, { "id": 63475, "author_name": "", "author_url": "", "post_date": "02/03/2015 10:07:25", "content": "

Is this dataset used for any other published paper in a journal or conference? Please, if you know, kindly refer me to it. 

", "votes": null, "replies": [] }, { "id": 63478, "author_name": "emanuele", "author_url": "", "post_date": "02/03/2015 10:42:52", "content": "

Hi Fatima,

\n

Dan Wakeman and Rik Henson, the authors of the original dataset from which the one of the competition was extracted, recently published this paper: http://www.nature.com/articles/sdata20151

\n

Additionally, last December we prepared a short article for the Machine Learning In NeuroImaging (MLINI) workshop of NIPS (http://nips.cc/Conferences/2014/Program/event.php?ID=4307). There, a portion of the competition dataset was used to test a proposed algorithm. The paper was accepted and it is expected to be published in the workshop's proceedings in 2015.Best,Emanuele

", "votes": null, "replies": [] }, { "id": 63555, "author_name": "", "author_url": "", "post_date": "02/04/2015 05:54:13", "content": "

Thank you for the quick reply. It helped me alot. 

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "47340": "", "47341": "", "47342": "", "47349": "", "47351": "", "63475": "", "63478": "", "63555": "" }, "source": "meta" }