The mailing list for listmasters using Sympa

[address@concealed]

Hi

May I insist about a request from Jose Marcio Martins :

Jose  send a message a few days ago ago asking for a corpus of messages 
for an antispam tool. He is looking for a public archive of lists 
without spam.  He need it for j-chkmail an spam filter server which use 
may diffrent technics including bayesien content filtering.

Thoses filters needs to be initialized with a set of spam and a set of 
ham. Spam is easy to catch, but getting ham is not so easy. Public list 
archive can be used for that. I allready send to Jose some but they are 
mainly french speacking lists or technical lists.

I known that Jose will be really very care with those messages (he parse 
them and then remove them), only statistics tokens are stored, so you 
can send him a tgz of some "arctxt" directories without any problems (he 
claim only for public list archives).

If you have some please contact address@concealed

Serge Aumont



> This is out of topic, sorry, but it seems to me that maybe some of you
> can help me.
>
> For the needs of my work on spam filters, I need to constitute a corpus
> of legitimate messages in a language other than french. For the moment,
> I'm interested in english messages.
>
> I can subscribe to some public discussion lists (and I've already done
> for cnn and foxnews newsletters), but this will take some time to
> accumulate a number of messages large enough.
>
> So, I'm looking for archives of public discussion lists. If some of you
> (who administer listservers) have this available and can pass it to me,
> please contact me privately.
>
> These messages will be used only for research purposes and won't be
> redistributed.
>
> Thanks,
>
> Jose-Marcio