Message-ID: <50D89B2F.6050102@sapo.pt>
Date: 2012-12-24T18:13:03Z
From: Rui Barradas
Subject: Spammer radhi
In-Reply-To: <loom.20121224T165405-117@post.gmane.org>
Hello,
I believe that today's spam filters already use statistical tools,
namely, non-parametric bayesian methods. This would mean that they can
adapt to known attacks but are useless against new ones.
Merry Christmas,
Rui Barradas
Em 24-12-2012 15:57, Ben Bolker escreveu:
> Arun Kumar Saha <arun.kumar.saha <at> gmail.com> writes:
>
>> Can the Statisticians here develop some good Statistical tool to stop this
>> Spamming keeping Type-I error almost zero?
> As Bruce Schneier (surely among others) has pointed out, the difference
> between security against natural risks (analogous to most statistical
> problems) and security against attackers is that the attackers adapt
> to avoid countermeasures. In the terminology of Hurlbert (1984), a classic
> in ecological statistics, this is the difference between "non-demonic"
> and "demonic" interventions; the tools of statistics are generally intended
> to guard against non-demonic intervention only.
>
> Ben Bolker
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.