(343) previous ~ index ~ next
To: tdt-distrib@unagi.cis.upenn.edu
From: James Allan <allan@cs.umass.edu>
Subject: Re: objections to including topic affiliation in FSD output
Date: Mon, 19 Nov 2001 13:35:09 -0500
> > 6. If, in addition to standard FSD (or NED as per Charles's renaming)
> > we want to score also 2nd or 3rd stories (mistaken for first story)
> > with declining value, we certainly can do so.
>
> This is not correct. Without a topic assignment, there is no way of
> knowing which stories are first, second or third.
I do not see any problem with asking systems to provide topic IDs in
their NED/FSD output. However, if we consider NED/FSD an interesting
and important problem on its own right, then we should not *force* it
to become just a variation on topic detection. There are techniques
that can be used that do not depend on topic threading, or that do not
depend on precise and careful topic threading. Forcing people away
from that limits technical options and is a mistake for a program such
as this one, where bold new ideas are what we're calling for in the
next year.
The NED/FSD evaluation task force (Jaime, Rich, and George?) should
come up with a proposed evaluation scheme and then float it to the
list rather than debating it in an open forum. The scheme should
support an evaluation that does not require topic IDs, and it should
also consider evaluation options that utilize topic threads of they're
available.
-- james
-------------------------------------------------------------
To unsubscribe from tdt-distrib, email majordomo@ldc.upenn.edu
with "unsubscribe tdt-distrib" in the body of the message.
(343) previous ~ index ~ next
Last updated Mon Nov 19 13:57:07 2001