[tv14.list] SIN - change in system training types
Paul Over
over at nist.gov
Fri Mar 14 08:45:01 EDT 2014
Hi,
For participants in the semantic indexing task ....
We have made a change in the interpretation of the system
training types to reflect the increasing importance of training
data choice on a par with choice of algorithms.
In choosing the correct category, ALL data used for training at
any level of any system component should be considered. This means
that even just the use of something like a face detector that was
trained on non-IACC training data would DISqualify the run as type
"A". This implies that some systems accepted in category A in the
previous years will be placed in categories B, C or D with the new
and more strict rules.
While the categories will be taken more strictly than in the previous
years, they will be used only as an information for clarifying what is
done by the participants. They will not be used for presenting the
results in different tables and figures; there will be only one global
ranking and one global plot in which all systems will be gathered, they
will be just tagged (as previously) by the category as part of the run
named generated at NIST
The guidelines for notebooks papers - the structured abstract - have
been changed to include a section for listing training data types.
Please take a look at these parts of the Guidelines:
http://www-nlpir.nist.gov/projects/tv2014/restrictions.html
http://www-nlpir.nist.gov/projects/tv2014/notebookpapers.html
Thanks,
Paul
--
Paul Over
----------------------------------------------
Retrieval Group
Information Access Division
Information Technology Laboratory
National Institute of Standards and Technology
Gaithersburg, MD 20899-8940 USA
Bldg. 225 Rm. A211 (Mailstop 8940)
Voice: 301 975-6784 Fax: 301 975-5287
More information about the tv14.list
mailing list