<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=iso-8859-1" http-equiv=Content-Type>
<META content="MSHTML 5.00.2920.0" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV style="FONT: 10pt arial"></DIV>
<DIV><FONT face=Arial size=2>Some days ago,</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>I asked (on several lists) about tools and info on
vectors</FONT><FONT face=Arial size=2> distance and</FONT><FONT face=Arial
size=2></FONT></DIV>
<DIV><FONT face=Arial size=2>indexing strategies. My question was very
general, however </FONT><FONT face=Arial size=2>the main
target</FONT></DIV>
<DIV><FONT face=Arial size=2>was concerned with IR application. I was
expecting</FONT><FONT face=Arial size=2> answers about</FONT><FONT face=Arial
size=2> packages</FONT><FONT face=Arial size=2></FONT></DIV>
<DIV><FONT face=Arial size=2>for computing any kind of features distances
(vectors,</FONT><FONT face=Arial size=2> Boolean, </FONT><FONT face=Arial
size=2>Euclide,</FONT></DIV>
<DIV><FONT face=Arial size=2>Levenshtein...). I should have said that our system
implements its own</FONT></DIV>
<DIV><FONT face=Arial size=2>indexing</FONT><FONT face=Arial size=2>
strateg</FONT><FONT face=Arial size=2>y.</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>I would like to thanks:</FONT></DIV>
<DIV><FONT face=Arial size=2>Romaric Besancon, Eric Gaussier, Paul
Holmes-Higgin,</FONT></DIV>
<DIV><FONT face=Arial size=2>Andrew MacFarlane, Ian Soboroff, Richard
Boulton,</FONT></DIV>
<DIV><FONT face=Arial size=2>Jian-Yun Nie,</FONT><FONT face=Arial size=2> and
Christian Boitet. </FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Here is a survey of the available
tools:</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Andrew McCallum's Bag Of Words
library:</FONT></DIV>
<DIV><FONT face=Arial size=2>Open source, seems complete.</FONT></DIV>
<DIV><FONT face=Arial size=2><A
href="http://www.cs.cmu.edu/~mccallum/bow">http://www.cs.cmu.edu/~mccallum/bow</A>
</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>SMART: it is a very complete IR system (indexing,
retrieval,</FONT></DIV>
<DIV><FONT face=Arial size=2>stop words for English and
Spanish...),</FONT></DIV>
<DIV><FONT face=Arial size=2>totally open source.</FONT></DIV>
<DIV><FONT face=Arial size=2>(<A
href="ftp://ftp.cs.cornell.edu/pub/smart/)">ftp.cs.cornell.edu/pub/smart/)</A>.
</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Muscat:</FONT></DIV>
<DIV><FONT face=Arial size=2><A
href="http://open.muscat.com/">http://open.muscat.com/</A>
</FONT></DIV>
<DIV><FONT face=Arial size=2>The indexing portion of Muscat is still
closed-source.</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>I have started to install SMART.</FONT></DIV>
<DIV><FONT face=Arial size=2>Thanks again,</FONT></DIV>
<DIV><FONT face=Arial size=2>Patrick</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>__________________________________<BR>Patrick
Ruch<BR>HUG - Medical Informatics Division<BR>CH-1211 Geneva 14<BR>tel.: (+41
22) 372 61 64<BR>fax: (+41 22) 372 48 55<BR>email: <A
href="mailto:Patrick.Ruch@dim.hcuge.ch">Patrick.Ruch@dim.hcuge.ch</A></FONT></DIV></BODY></HTML>