Se connecter

Ou se connecter avec
Date limite de participation :
15 juin 2015

Data Science Game

Welcome to the 1st edition of the Data Science Game!
www.datasciencegame.com

Classement
1. (1) MInd overflow Score 98,19713%
2. (2) Insight UCD Score 91,82019%
3. (3) Telecominers - Score 89,89923%
Ce challenge est terminé.

Projet pédagogique

607

contributions

20

participants

terminé
terminé
For the 1st stage of the competition you are given excerpts of books and asked to guess the names of their writers among William Shakespeare, Mark Twain, Oscar Wilde, Edgar Allan Poe, Jane Austen, and Arthur Conan Doyle.

The training set is simply constituted with 2 columns containing the excerpts and their matching author. You are asked to use this training set to design methods and algorithms allowing you to guess the relevant authors.

The efficiency of your methods will be compared using a test set: the score for each submission will just be the rate of good answers.

Even if the contest is friendly each team should provide minimum 3 trials:

the first one before the 1st of June, 10 am French time
the second one before the 8th of June, 10 am French time
the last one before the 15th of June, 10 am French time
The friendly contest will be closed on the 15th of June at 10 am French time. 

For this 1st stage you are not limited for the number of submissions.

A friendly ranking will be daily updated.

          
For the competition each team will be allowed to use the computational resources of its choice. Powerful laptops equipped with the softwares you are used to should be sufficient.

Nevertheless, in order to ensure the fairness of the competition, each team will benefit of an access to the powerful Cloud Platform of Google (https://cloud.google.com/) where
  • you will be able to install the relevant free softwares of your choice,
  • you will be able to use a maximum of $500 for computational queries during the two phases.
As a complement you are free to use the free trial offered by the Cloud Platform of Google. 

Good luck!

You must submit a file in CSV format, using semi-column as separator, with 1 header and 2 columns:

  • Column 1: line Id
  • Column 2: author surname


File example:

Id;Author
1;austen
2;twain
......

The score will be the rate of good answers.

1. (1) MInd overflow 30 contributions 14/06/15 17:32 Score 98,19713%
2. (2) Insight UCD 14 contributions 13/06/15 22:54 Score 91,82019%
3. (3) Telecominers - 66 contributions 13/06/15 22:17 Score 89,89923%
4. (4) Dmitry Ulyanov 36 contributions 12/06/15 18:31 Score 89,47410%
5. (5) MADatascience BRO 97 contributions 15/06/15 22:48 Score 89,27728%
6. (6) ctech IITD 66 contributions 14/06/15 22:17 Score 88,56086%
7. (7) Poly Unicorns 36 contributions 09/06/15 15:19 Score 88,45064%
8. (8) Nedap Hinton Hyenas 26 contributions 10/06/15 14:29 Score 88,26169%
9. (9) Sapienza 52 contributions 11/06/15 00:27 Score 88,00189%
10. (10) Breaking Data 2 contributions 03/06/15 18:39 Score 86,61628%
11. (12) Full Metal 14 contributions 15/06/15 00:06 Score 85,79751%
12. (11) TSE 2 contributions 31/05/15 19:55 Score 85,78177%
13. (13) GMM INSA 19 contributions 11/06/15 15:47 Score 85,08896%
14. (14) Designated Neurons 26 contributions 19/05/15 06:50 Score 82,69564%
15. (15) Ayrbus-Fishmanati Ayrbus-Fishmanati 7 contributions 01/06/15 09:54 Score 82,69564%
16. (16) SID Team 9 contributions 26/05/15 23:05 Score 80,68021%
17. (17) dsg eNStA 6 contributions 13/06/15 17:21 Score 78,39710%
18. (18) A Giant Mind 6 contributions 08/06/15 10:50 Score 76,63360%
19. (19) Cocotte Data 8 contributions 01/06/15 00:12 Score 76,43678%
20. (20) UniMAnalytics Benjamin Schäfer 4 contributions 01/06/15 13:00 Score 49,76382%
Discussions
loading... Chargement...