SHINRA2020-ML: How to Participate

 

Schedule

  • January 2020               Data release
  • August 31, 2020              Registration & Result submission deadline
  • mid September, 2020        Evaluation results due back to participants
  • December 2020          NTCIR-15 Conference (NII, Tokyo)

Just wanna peek: Willing to know if it’s interesting

If you want to see what it would be like, you can download (Trial Datasets“),  which includes training data.

SHINRA2020-ML aims to categorize 30 language Wikipedia entities into ENE (Extended Named Entity) categories. When you participate in the task, you can choose at least one target language. The training data will be used to categorize all the remaining Wikipedia entities of the language. 


Procedure of Full-Participation

  1. Registration through NTCIR-15
  2. Account Creation and Data download
  3. Result Submission
  4. Reporting

1.  Registration through NTCIR-15

SHINRA2020-ML is a task in NTCIR-15 of the NTCIR evaluation workshop series. If you would like to participate in the task, please register through the following page:

NTCIR-15 Registration page.

2.  Account Creation and Data Download

You need an account to get any of the data provided for the task. Please create your SHINRA account at:

SHINRA: Sign in page.

When you download the data, you have to login using the account.  

The training data (already categorized Wikipedia entities) and target data (the Wikipedia entities you will categorize) are available from SHINRA 2020-ml Data download page. 

3. Result Submission

Please submit the data through SHINRA2020-ML: Results Submission page.

4.  Reporting

Please submit a “system description report“. 
The results can be presented at the NTCIT-15 conference. Further details will be announced later.