LINGUIST List 18.869
|
Thu Mar 22 2007
Confs: Computational Ling,Dravidian Lang,Indo-Iranian Lang/India
Editor for this issue: Jeremy Taylor
<jeremy linguistlist.org>
|
To post to LINGUIST, use our convenient web form at
http://linguistlist.org/LL/posttolinguist.html.
|
Directory
1. Anil Kumar
Singh,
Named Entities Contest for South Asian Languages
Message 1: Named Entities Contest for South Asian Languages
|
Date: 20-Mar-2007
From: Anil Kumar Singh <anilresearch.iiit.net>
Subject: Named Entities Contest for South Asian Languages
Named Entities Contest for South Asian Languages
Short Title: NLPAI-MLContest
Date: 07-Jul-2007 - 07-Jul-2007
Location: Hyderabad, A.P., India
Contact: Anil Kumar Singh
Contact Email: anilresearch.iiit.net
Meeting URL: http://ltrc.iiit.net/nlpai_contest07/
Linguistic Field(s): Computational Linguistics
Language Family(ies): Dravidian; Indo-Iranian
Meeting Description:
Machine learning techniques are now an integral part of most Natural Language
Processing (NLP) systems. NLPAI Machine Learning Contest focuses on application
of such techniques for one major NLP problem every year. This time the contest
topic is Named Entity Recognition for South Asian languages.
NLPAI Machine Learning Contest 2007
Host: IIIT, Hyderabad, India
Shared Task : Named Entity Recognition (For South Asian Languages)
Competition
The contest is open for systems which can automatically annotate text with named
entities for South Asian languages. As part of the contest, the participants
will be provided with named entity annotated data in some South Asian languages.
Initial data will be made available to the contestants to improve and refine
their systems. For the final contest, a new data set will be provided. Selected
contestants will be invited to present their techniques and results at a workshop.
The contestants can (optionally) annotate new data with the given tag set and
submit for the contest. Those who provide such data will receive more data for
training their systems than the other contestants.
There will be two separate tracks:
1. Closed Track (track-I): Contestants can only use the resources allowed as
mentioned below
2. Open Track (track-II): Contestants can also present the results obtained with
any other resources (or without any resources), in addition to the results for
the shared task for track-I
Resources
- Available implementations of machine learning techniques
- Annotated data provided for the task
- Extra annotated data submitted the participants will be available to all the
participants who submit such data
Certificates and Prizes
The three best systems would be given certificates and cash awards.
1st prize: Rs. 7,000
2nd prize: Rs. 4,000
3rd prize: Rs. 2,500
The shortlisted systems would be invited to present their systems at the
workshop (July, 2007).
Important Dates
Announcement of the contest: 20 March
Registration for the contest closes: 15 April
Release of initial annotated data (training and development): 20 April
Annotated data submission by participants (optional): 15 May
Trial submission based on training and development data (for feedback):
15 May
Release of larger amount of data (only to those contestants who submit annotated
data):
1 June
The first round of the contest (submission for review): 10 June
Submission of write-ups (on approach used): 15 June
Notification of acceptance: 25 June
The final round (submission for presentation) and the workshop: 7 July
Contest Website
http://ltrc.iiit.net/nlpai_contest07/
A half-day tutorial will be conducted on 7 April at IIIT, Hyderabad, India to
provide some guidance about the task, the required tools and the resources to be
used.
Event Coordinator
Anil Kumar Singh
Language Technologies Research Centre
International Institute of Information Technology
Hyderabad, India
Respond to list|Read more issues|LINGUIST home page|Top of issue
|
|
Please report any bad links or misclassified data
LINGUIST Homepage | Read
LINGUIST | Contact us
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.
|
|
