[LFG] FinTOC’2 shared task: release of training data

[FinTOC SharedTask]

FinTOC’2 shared task



News: The training data has been released. If you wish to access it, you
need to register to the shared task here:
https://forms.gle/LFsVaw6DqYikhKHx9




Held at COLING 2020 as part of the FNP-FNS 2020 workshop.

13 September, Barcelona, Spain.

====================


Shared Task URL: http://wp.lancs.ac.uk/cfie/fintoc2020/

Workshop URL: http://wp.lancs.ac.uk/cfie/fnp2020/

Participation Form: https://forms.gle/LFsVaw6DqYikhKHx9

_____________________________________________

The FinTOC’2 shared task aims to bring together the community of
researchers interested in Financial Document Processing and Document Layout
Analysis to advance the state of the art in the automatic processing of
financial documents. This task focuses on the automatic generation of
reports' Table Of Contents (henceforth TOC), as it is a key building block
in the semantic analysis of financial documents. Generating the TOC
requires detecting the span of all document sections and subsections,
identifying their titles, and organising them into a hierarchy. It is a
well-known fact that extracting document structure is a key step in
information processing. For example sections can be used to determine areas
where algorithms can be applied, such as Information Extraction, thus
reducing false positives rate and irrelevant noise.

This is the second edition of the FinTOC shared task which will be held at
COLING 2020 in Barcelona (Spain) as part of the FNP-FNS 2020 workshop. Last
year’s edition received significant interest, particularly on the Title
Detection track. Our aim this year is to increase interest by:

- lowering the barriers to the entry to the TOC extraction track, and

- opening up the task to a new language: French. We are particularly
interested in systems which can be applied to both English and French
languages.

This second edition proposes two tracks: one track per language, and it
will score systems on both Title detection and TOC generation performance.
We have revised the task and greatly simplified data formats to make it as
smooth as possible for every interested researcher to participate and
submit their systems’ outputs at FinTOC’2.

Each of the participating teams will be asked to submit a short paper
describing their methods and solutions to be presented at the workshop.

_____________________________________________

To register your interest in participating in FinTOC’2 shared task please
use the following google form by no later than April 6th, 2020:
https://forms.gle/LFsVaw6DqYikhKHx9

Soon after, you will receive a link to download the training data.

__________________________________________


Important dates:

December 1st, 2020: Registration opens.

February 17th, 2020: Release of training set & scoring scripts.

March 23rd, 2020: Release of test set.

April 6th, 2020: Registration deadline.

April 13th, Submission deadline.

May 1st, 2020: Release of results.

Sep 13th, 2020: Workshop day.

_________________________________________

Contact:

For any questions on the shared task please contact us on:

fin.toc.task at gmail.com

______________________________________


Shared task organizers:

- Najah-Imane Bentabet, Fortia Financial Solutions

- Ismail El Maarouf, Fortia Financial Solutions

- Mahmoud El-Haj, Lancaster University

- Remi Juge, Fortia Financial Solutions

- Dialekti Valsamou-Stanislawski, Fortia Financial Solutions

- Virginie Mouilleron, Fortia Financial Solutions
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lfg/attachments/20200219/579f6278/attachment.htm>