I don't understand the thread.
If you want to take any special actions on Hebrew content, use the regex. 
Give it any special score, or white or np or spamlover .... or what ever.

>> I worry about not having enough HM / Bayes statistics on hebrew and I'd
>> hate to reduce the amount of English data if we were to start gathering
>> Hebrew for this one user.

Just think about this! 

Who knows what are "enough HM / Bayes statistics on hebrew" ?
How would anyone "reduce the amount of English data" ? And WHY???

DON'T expect an answer on these questions from me.

If you want or need a more or less 'privat' heuristic of Bayes/HMM - set 
'DoPrivatSpamdb' accordingly.

ASSP (Bayes/HMM) never cares about a language (unicode script or block), 
except for word stemming.

For example:
until now , the sense (in english) of "(we offer | I'm looking for)  nice 
used german cars" has commonly a very high HMM spam weight.
now, the Hebrew speeking user is looking for such a car using also Hebrew 
in the mails. The first mail would be possibly blocked - the user reports 
HAM - now assp knows that mails with "(we offer | I'm looking for)  nice 
used german cars" and Hebrew in it, from or to this user are HAM - even 
there is only one mail reported  There is nothing changed for all other 
users, related to such mails. Only the privat ham weight (related tor this 
sense) for this user will change.

Thomas





Von:    K Post <nntp.p...@gmail.com>
An:     ASSP development mailing list <assp-test@lists.sourceforge.net>
Datum:  12.01.2016 18:32
Betreff:        Re: [Assp-test] Multiple languages / charactersets



Excellent.  Of course, now I start worrying that any email with a Hebrew
character in it will bypass Bayes/HM...  (I know I asked for this, just am
second guessing that this is a good idea)...

On Mon, Jan 11, 2016 at 11:32 AM, Thomas Eckardt 
<thomas.ecka...@thockar.com
> wrote:

> >and that would be all that's needed?
>
> Yes, I was hoping so. But the simple unicode script match does currently
> not work for several good reasons.
> But you can use the following regex - it contains each character of the
> hebrew alphabet, which is nearly the same like \p{Hebrew}+
>
> [אבגדהוזחטיכךלמםנןסעפףצץקרשת]+
>
> Thomas
>
>
>
> Von:    K Post <nntp.p...@gmail.com>
> An:     ASSP development mailing list <assp-test@lists.sourceforge.net>
> Datum:  11.01.2016 02:24
> Betreff:        Re: [Assp-test] Multiple languages / charactersets
>
>
>
> Interesting!
>
> Does this mean that I simply put Hebrew}
>
> 1) in baysSpamLoversRE so that other filters will apply but we won't
> bother
> with HMM/Bayes checks on inbound mail
> and
> 2) in RedRE to stop any email that contains a hebrew character from 
being
> added to the corpus on send
>
> and that would be all that's needed?
>
> On Fri, Jan 8, 2016 at 11:54 AM, Thomas Eckardt
> <thomas.ecka...@thockar.com>
> wrote:
>
> > Unicode will help
> >
> > the regular expression:
> >
> > \p{Hebrew}
> >
> > for example, will detect a singe hebrew character
> >
> > Thomas
> >
> >
> >
> > Von:    K Post <nntp.p...@gmail.com>
> > An:     ASSP development mailing list 
<assp-test@lists.sourceforge.net>
> > Datum:  08.01.2016 17:36
> > Betreff:        [Assp-test] Multiple languages / charactersets
> >
> >
> >
> > One of our users is bilingual and apparently switches between English
> and
> > Hebrew character set frequently  This is the only user like this that 
I
> > know of.
> >
> > I worry about not having enough HM / Bayes statistics on hebrew and 
I'd
> > hate to reduce the amount of English data if we were to start 
gathering
> > Hebrew for this one user.
> >
> > Is there a way to tell ASSP to essentially redlist and no-process mail
> > send
> > with the hebrew characters in the email (even if it's a mixed 
message)?
> > That way his english emails would enjoy the same antispam protection 
as
> > the
> > other users.  If not, I guess I need to redlist and no-process (or no
> > collect, bayes lover, hm lover) his email address if I want to keep
> hebrew
> > out?
> >
> > Is there a better way of thinking about this?
> >
> > Thanks
> > Ken
> >
> >
>
> 
------------------------------------------------------------------------------
> > _______________________________________________
> > Assp-test mailing list
> > Assp-test@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/assp-test
> >
> >
> >
> >
> > DISCLAIMER:
> > *******************************************************
> > This email and any files transmitted with it may be confidential,
> legally
> > privileged and protected in law and are intended solely for the use of
> the
> >
> > individual to whom it is addressed.
> > This email was multiple times scanned for viruses. There should be no
> > known virus in this email!
> > *******************************************************
> >
> >
> >
> >
>
> 
------------------------------------------------------------------------------
> >
> > _______________________________________________
> > Assp-test mailing list
> > Assp-test@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/assp-test
> >
> >
>
> 
------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
> _______________________________________________
> Assp-test mailing list
> Assp-test@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/assp-test
>
>
>
>
> DISCLAIMER:
> *******************************************************
> This email and any files transmitted with it may be confidential, 
legally
> privileged and protected in law and are intended solely for the use of 
the
>
> individual to whom it is addressed.
> This email was multiple times scanned for viruses. There should be no
> known virus in this email!
> *******************************************************
>
>
>
>
> 
------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
> _______________________________________________
> Assp-test mailing list
> Assp-test@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/assp-test
>
>
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
_______________________________________________
Assp-test mailing list
Assp-test@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/assp-test




DISCLAIMER:
*******************************************************
This email and any files transmitted with it may be confidential, legally 
privileged and protected in law and are intended solely for the use of the 

individual to whom it is addressed.
This email was multiple times scanned for viruses. There should be no 
known virus in this email!
*******************************************************


------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
_______________________________________________
Assp-test mailing list
Assp-test@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/assp-test

Reply via email to