TIPSTER Volume 1

<p><a href="http://catalog.ldc.upenn.edu/LDC93T3A" rel="nofollow">LDC93T3A</a> - Complete TIPSTER corpus</p><br>
<p>LDC93T3B - Volume 1 of the TIPSTER corpus</p><br>
<p><a href="http://catalog.ldc.upenn.edu/LDC93T3C" rel="nofollow">LDC93T3C</a> - Volume 2 of the TIPSTER corpus</p><br>
<p><a href="http://catalog.ldc.upenn.edu/LDC93T3D" rel="nofollow">LDC93T3D</a> - Volume 3 of the TIPSTER corpus</p><br>
<p>TIPSTER 1 is sometimes also called the Text Research Collection Volume 1 or TREC-1.</p><br>
<p>The TIPSTER project was sponsored by the Software and Intelligent Systems Technology Office of the Advanced Research Projects Agency (ARPA/SISTO) in an effort to significantly advance the state of the art in effective document detection (information retrieval) and data extraction from large, real-world data collections.</p><br>
<p>The detection data is comprised of a test collection built at NIST for the TIPSTER project and the related TREC project. The TREC project has many other participating information retrieval research groups, working on the same task as the TIPSTER groups, but meeting once a year in a workshop to compare results (similar to MUC). The test collection consists of three CD-ROMs of SGML encoded documents distributed by LDC plus queries and answers (relevant documents) distributed by <a href="http://trec.nist.gov/" rel="nofollow">NIST</a>.</p><br>
<table border="1"><br>
<tbody><br>
<tr><br>
<td>Source</td><br>
<td>Year</td><br>
<td>Approx. # Words (Millions)</td><br>
</tr><br>
<tr><br>
<td>Associated Press</td><br>
<td>1989</td><br>
<td>40</td><br>
</tr><br>
<tr><br>
<td>Wall Street Journal</td><br>
<td>1987</td><br>
<td>20</td><br>
</tr><br>
<tr><br>
<td>Wall Street Journal</td><br>
<td>1988</td><br>
<td>17</td><br>
</tr><br>
<tr><br>
<td>Wall Street Journal</td><br>
<td>1989</td><br>
<td>6</td><br>
</tr><br>
<tr><br>
<td>Dept. of Energy</td><br>
<td>&nbsp;</td><br>
<td>28</td><br>
</tr><br>
<tr><br>
<td>Federal Register</td><br>
<td>1989</td><br>
<td>38</td><br>
</tr><br>
<tr><br>
<td>Ziff/Davis</td><br>
<td>&nbsp;</td><br>
<td>36</td><br>
</tr><br>
</tbody><br>
</table><br>
<p>&nbsp;</p><br>
<p>The documents in the test collection are varied in style, size and subject domain. The first disk contains material from the <a href="desc/addenda/LDC93T3B_WSJsample" rel="nofollow">Wall Street Journal</a> (1986, 1987, 1988, 1989), the <a href="desc/addenda/LDC93T3B_APsample" rel="nofollow">AP Newswire</a> (1989), the <a href="desc/addenda/LDC93T3B_FRsample" rel="nofollow">Federal Register</a> (1989), information from <a href="desc/addenda/LDC93T3B_CSsample" rel="nofollow">Computer Select</a> disks (Ziff-Davis Publishing) and short abstracts from the <a href="desc/addenda/LDC93T3B_DOEsample" rel="nofollow">Department of Energy</a>. The second disk contains information from the same sources, but from different years. The third disk contains more information from the Computer Select disks, plus material from the <a href="desc/addenda/LDC93T3D_SJMercurysample" rel="nofollow">San Jose Mercury News</a> (1991), more AP newswire (1990) and about 250 megabytes of formatted <a href="desc/addenda/LDC93T3D_USPatent" rel="nofollow">U.S. Patents</a>. The format of all the documents is relatively clean and easy to use, with SGML-like tags separating documents and document fields. There is no part-of-speech tagging or breakdown into individual sentences or paragraphs as the purpose of this collection is to test retrieval against real-world data.</p><br>
<p>The three Tipster discs so far released have been re-issued with updates and corrections and all recipients of the earlier versions should have received these replacements free of charge. If you think you have the unrevised original, contact LDC for confirmation.</p></br>