Me, too! :-)
We are currently playing with the small Reuters collection (about 21.500 news items from the 80s), but I don't know if I am allowed to distribute it and it is too small anyway -- many of the implications we find are based on 1 to 3 documents. I still have a collection of 5 CDs I downloaded from Google about 2 years ago, which I still haven't found the time to look at -- it probably will be one of the next things to do. I can see what their licence says about redistribution.
Another good source for collections is http://www.kdnuggets.com/datasets/index.html -- but I haven't found a good news corpus yet.
I am definitely interested in comparing some results on whatever collection we can find. I'll talk to our IR guys in the next days to see what they know about.
Peter
Jos van der Meer wrote:
For my experiments with Lucene, I would like to have a publicly available free, medium size, downloadable corpus of newspaper articles (topics do not matter, nor does its publication date).
For I would like to share the results of the experiments, and people should be able to reproduce and to extend it.
Don't send the corpora themselves (..), but please send me their URLs.
Thanks in advance,
[EMAIL PROTECTED] aidministrator nederland bv - http://www.aidministrator.nl/ prinses julianaplein 14-b, 3817 cs amersfoort, the netherlands tel. +31-(0)33-4659987 fax. +31-(0)33-4659987
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
