(203) previous ~ index ~ next

---

To: TDT Distrib <tdt-distrib@ldc.upenn.edu>
From: Jonathan Fiscus <jonathan.fiscus@nist.gov>
Subject: Error in the Story Link evaluation spec
Date: Thu, 30 Sep 1999 15:53:32 -0400

---

Folks,

Ralf Brown pointed out an error in the story link detection evaluation
spec., (V2.7), that wasn't broadcast to the mailing list. The error is
in section A.6, "Link Detection I/O Formats" under the "Output data"
heading. The field descriptions for "Story_ID_1" and "Story_ID_2"
should be identical to the index file descriptions of these fields. The
text should be:

Story_ID_1: The character string story identifier of the first story.
This identifier is the source file and TDT3 corpus docno, separated be a
colon.

Story_ID_2: The character string story identifier of the second story.
This identifier is the source file and TDT3 corpus docno, separated be a
colon.

Please use these identifiers if you submit story link detection results.

Thanks!
Jon


--
Jonathan Fiscus
National Inst. of Stds. and Tech.
100 Bureau Dr. Stop 8940
Gaithersburg, MD 20899-8940

Phone: (301) 975-3182
Email: jonathan.fiscus@nist.gov

---

(203) previous ~ index ~ next

Last updated Tue Oct 19 10:10:08 1999