Submission

This page includes two parts:

  1. Submission Instructions
  2. Submission Format

Submission Instructions

To submit your results, (1) register at BioASQ website and (2), follow the following instructions:

A. Up to 5 submissions per sub-track will be allowed.

B. You must submit ONE SINGLE ZIP file with the following structure:

  • One subdirectory per subtask in which you are participating. 
  • For track 2, separate the submissions in each language in different directories, using en for English, es for Spanish and it for Italian.
  • In addition, in the parent directory, you must add a README file with your contact details and a really short explanation of your system.
  • If you have more than one system, you can include their predictions and we will evaluate them (up to 5 prediction runs).
  • For each prediction run:
    • You must include a single tab-separated file (.TSV) with your predictions.
    • Your .TSV file must include headers (see Submission Instructions below to know which ones) 
    • If you have more than one system, include one tab-separated file for each system.
    • If you have more than one system, name the tab-separated files with numbers and a more or less recognizable name. For example, 1-systemDL.tsv and 2-systemlookup.tsv
    • Remember for explain each of them shortly in the README file!
  • This is an example of what your .ZIP file should look like:

C. Once your .ZIP file is ready, access the BioASQ’s Participants Area, click on Submitting > Task MultiCardioNER, where you will be able to upload your predictions. Keep in mind that only one .ZIP file is allowed. Uploading a new one will replace the previous one.

Submission Instructions

All submissions must be done using .TSV files. Both tracks use the same format.

  • filename: Name of the file from which the mention has been extracted.
  • label: Label assigned to the extracted mention. It should be ENFERMEDAD for Track 1 or FARMACO for Track 2.
  • start_span: Character number where the detected mention starts.
  • end_span: Character number where the detected mention ends.
  • text: Mention extracted from text