Re: CRM114 as a 1-way classifier

On Thu, Aug 06, 2009 at 04:15:24PM -0400, MightyByte spake thusly:
> I definitely understand this problem, but it's still possible to come
> up with with an idea of what's normal and what's not without having a
> human-specified categorization.  Z-scores in statistics do just that.

If you ever come up with a working implementation I would love to see
it. I wrote a logfile analysis tool in Python which works well for me
so far but it isn't terribly scalable and requires me to put in a
regex (generated by the tool itself with appropriate wildcards etc
inserted for routinely changing values like PIDs, IPs, etc. so it is
just cut and paste for me with occasional modification) into an ignore
file. I used to have a website up for this tool on my personal webpage
but it seems to have gotten lost. I'll have to get it back up again soon...

-- 
Tracy Reed
http://tracyreed.org

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july

_______________________________________________
Crm114-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/crm114-general

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)

iD8DBQFKe1Fx9PIYKZYVAq0RAkaDAKCEUzjz37RaNE74INcdQiH5zvaVCgCfVTgM
O4HUog+IsAM02oVFM0HIfkM=
=3SZG
-----END PGP SIGNATURE-----