[Assp-test] Question about spam reporting / language detection

[Dirk Kulmsee]

Hi everybody,

I am currently running ASSP version 2.4.4(14307) on Perl 5.20. As for spam
reporting through the email interface I understand it is best to send the
complete mail as an attachment so the headers are still available for
analysis. I think on my end the language detection does not work correctly.

When I report spam through the email interface I get this result:
-------------------------------------
Unicode Analysis:
the following non symbolic unicode blocks (except InBasicLatin) were found:
the following symbolic unicode blocks were found:
the following unicode scripts were found except(Common + Latin):

Bayesian Analysis: - word stemming engine is used - language swedish
detected
Bad Words       Bad Prob        Good Words      Good Prob
&#x72&#x65&#x70&#x6f&#x72&#x74&#x20&#x73&#x70&#x61&#x6d 0.9999           
&#x73&#x70&#x61&#x6d&#x20&#x61&#x73&#x73&#x70   0.9999           
                &#x66&#x72&#x69&#x20&#x6e&#x6f&#x76     0.0039
                &#x6e&#x6f&#x76&#x20&#x72&#x65&#x63&#x65&#x69&#x76&#x65&#x64
0.0152
&#x73&#x73&#x75&#x62&#x20&#x73&#x74&#x65&#x6c&#x6c&#x65&#x6e&#x61&#x75&#x73&
#x73&#x63&#x68&#x72&#x65&#x69&#x62&#x75&#x6e&#x67       0.9848           
                &#x6e&#x6f&#x76&#x20&#x63&#x65&#x74     0.0556
&#x66&#x72&#x6f&#x6d&#x20&#x6f&#x75&#x74        0.9444          
-------------------------------------

When I paste the same mail into the Mail Analyzer in the GUI I get this
(much better) result:
-------------------------------------
Unicode Analysis:
the following non symbolic unicode blocks (except InBasicLatin) were found:
the following symbolic unicode blocks were found:
the following unicode scripts were found except(Common + Latin):

Bayesian Analysis: - word stemming engine is used - language german detected

Bad Words       Bad Prob        Good Words      Good Prob
uns firma       1.0000           
uns team        1.0000           
die bezahl      0.9975           
uk sender       0.9972           
uk ssub         0.9956           
stund sie       0.9919          
-------------------------------------

Can anyone tell me what I would need to check and adjust?

Thanks a lot

Dirk



------------------------------------------------------------------------------
Comprehensive Server Monitoring with Site24x7.
Monitor 10 servers for $9/Month.
Get alerted through email, SMS, voice calls or mobile push notifications.
Take corrective actions from your mobile device.
http://pubads.g.doubleclick.net/gampad/clk?id=154624111&iu=/4140/ostg.clktrk
_______________________________________________
Assp-test mailing list
Assp-test@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/assp-test