Spam Filter ISP Support Forum

  New Posts New Posts RSS Feed - Bayesian Filer Test Not working
  FAQ FAQ  Forum Search   Register Register  Login Login

Bayesian Filer Test Not working

 Post Reply Post Reply Page  <12
Author
Sundance View Drop Down
Newbie
Newbie


Joined: 18 July 2006
Location: Hungary
Status: Offline
Points: 10
Post Options Post Options   Thanks (0) Thanks(0)   Quote Sundance Quote  Post ReplyReply Direct Link To This Post Posted: 13 September 2006 at 7:08am

I sent it again. Both the corpus and the logfiles, taht You asked for. (support@logsat.com)

Did you get it?

Sundance

Back to Top
LogSat View Drop Down
Admin Group
Admin Group
Avatar

Joined: 25 January 2005
Location: United States
Status: Offline
Points: 4068
Post Options Post Options   Thanks (0) Thanks(0)   Quote LogSat Quote  Post ReplyReply Direct Link To This Post Posted: 13 September 2006 at 9:58am
Got them. Please give us a few hours to analyze them.
Roberto Franceschetti

LogSat Software

Spam Filter ISP
Back to Top
Sundance View Drop Down
Newbie
Newbie


Joined: 18 July 2006
Location: Hungary
Status: Offline
Points: 10
Post Options Post Options   Thanks (0) Thanks(0)   Quote Sundance Quote  Post ReplyReply Direct Link To This Post Posted: 27 September 2006 at 2:55am
Any results?????
Back to Top
LogSat View Drop Down
Admin Group
Admin Group
Avatar

Joined: 25 January 2005
Location: United States
Status: Offline
Points: 4068
Post Options Post Options   Thanks (0) Thanks(0)   Quote LogSat Quote  Post ReplyReply Direct Link To This Post Posted: 27 September 2006 at 10:18pm
I'm sorry. Even the 2nd time we placed your corpus files in our own live environment, it was populated correctly. It does seem that your type of email traffic is not able to populate the corpus with meaningful statistical values. It is indeed strange that since most of your clean emails are in hungarian, while the spam is in english, that this is not enough to create valid statistical data. However this is the what the data in your database shows.

As a last resport, could you send us a few (a dozen or so) clean emails (in hungarian) that you received? Please note we'll need the original, unmodified email source, so we can try to see if the foreign language is causing any problems.

Roberto Franceschetti

LogSat Software

Spam Filter ISP
Back to Top
LogSat View Drop Down
Admin Group
Admin Group
Avatar

Joined: 25 January 2005
Location: United States
Status: Offline
Points: 4068
Post Options Post Options   Thanks (0) Thanks(0)   Quote LogSat Quote  Post ReplyReply Direct Link To This Post Posted: 28 September 2006 at 7:47pm
I think we may have indeed found something!

The accented words (ex kszl) are being incorrectly transformed by the Bayesian tokenization process. The characters with accents are being replaced by spaces, and this is definetly a problem, as will cause incorrect statistics about words.

Allow us a few more hours to ensure the fix we have in mind will not cause other issues. We will have a patched version shortly.

Thanks for your patience with all of this.

Roberto Franceschetti

LogSat Software

Spam Filter ISP
Back to Top
 Post Reply Post Reply Page  <12
  Share Topic   

Forum Jump Forum Permissions View Drop Down



This page was generated in 0.063 seconds.