Can't remember if this was posted here before or not, but my 
general understanding is that profiling as a general rule is stupid 
assuming that your terrorist group is smart enough to realize that 
profiling is going on and has enough individuals willing to sacrifice 
themselves that they have several options.  This paper describes why 
completely random "full screenings" are more effective than profiled ones 
when using any characteristics that remain constant over time to determine 
whether or not to profile.  The argument is essentially that you send a 
person on a flight several times before they plan to carry an explosive 
and if they get selected for a full screening most of the time, then you 
don't use them to carry an explosive on to the flight, since past 
searches are predictive of future searches.  This seems to indicate that 
profiling based on ethnicity would be useless given that the terrorists 
have people of different ethinicities they can choose from.  This paper 
was specifically written to comment on CAPS but could be applicable to a 
wider range of profiling systems.

http://www-swiss.ai.mit.edu/6805/student-papers/spring02-papers/caps.h...

	However, behavioral profiling is a new concept to me and seems 
like a different concept.  After all past behavior may not be the same as 
behavior when carrying an explosive and thus the carnival booth algorihtm 
can not be used.  Their own policy reccomendations in this paper seem to 
indicate that behavior based profiling would be the way to go, and I can 
agree with them on that.
Take care,
Bill

On Sun, 13 Aug 2006, Daniel Reeves wrote:

> Yeah, it's just like that in the states, except instead of enforcement 
> they're using the honor system.  It really is absurd.
> 
> Have you flown in Britain yet, James?  My mom was under the same 
> impression about the situation in the states.  I can tell you it's not the 
> case in New York.  The restrictions truly are a charade.
> 
> The aggressive restrictions you describe would be a hurdle for terrorists, 
> I suppose, but even then, simple circumventions like ziplock bags in 
> jacket linings would get past the pad-down searches.	Not to mention that 
> they carefully avoid your crotch in those searches so really you could 
> stick a bottle of something in your underwear and get through.
> 
> As to your point about diluting security by not profiling, there's one 
> other counter argument besides "profiling feels icky":  profiling changes 
> the hurdle from getting your bomb/knife/etc through diluted security to 
> getting your bomb/knife/etc into the luggage of Grandma Caucasian.  If the 
> latter is much harder (and apparantly the Israeli government thinks it is) 
> then you're right.
> 
> 
> --- \/   FROM James W Mickens AT 06.08.13 18:32 (Today)   \/ ---
> 
> > Israel has been dealing with the threat of airborne terrorism for decades, 
> > and their air transport system has an excellent track record of detecting 
> > potential terrorist attacks before they occur. Based on Israeli experience,

> > it appears that the best way to detect terrorists that are trying to board 
> > airplanes is to behaviorally profile them. Even for a highly indoctrinated 
> > terrorist, it's difficult to remain calm during the minutes before the 
> > terrorist act is committed. Law officers can be trained to detect this 
> > nervousness and then nominate the person for further screening. Duane
Woerth, 
> > president of the Air Line Pilots Association, provided the following 
> > testimony to the United States Congress: "The Israeli aviation 
> > security-screening model, widely regarded as the world's best, is 
> > human-centered and trust-based. Information is collected on passengers
before 
> > they arrive to the airport and they are physically screened and queried in 
> > concert with that knowledge. Trained personnel assess individual 
> > characteristics that are indicative of deception and engage passengers in 
> > conversation and questioning to establish the purpose and authenticity of
an 
> > individual's travel plans. Considerably less time and resources are spent
on 
> > physically screening those who are deemed to be non-threat persons and 
> > traveling for legitimate purposes." Woerth continued by saying that US 
> > airports erroneously assume "that everyone poses a potential threat to 
> > aviation security. The truth is that the vast majority of individuals, 
> > including airline pilots, do not pose any kind of threat to aviation . . . 
> > our screening resources are greatly diluted by giving the same degree of 
> > physical security to an Air Force Reserve general and airline pilot as is 
> > given to a federal prison parolee." Some limited forms of behavioral 
> > profiling have begun to appear in American airports, but not to the extent 
> > that it has been used in Israel. There are several reasons for this.
Perhaps 
> > most important is that the Israelis perform behavioral profiling in concert

> > with ethnic profiling, paying more attention to people who look Arabic.
This 
> > racial profiling is inherent to the Israeli notion that, probabilistically 
> > speaking, different types of people pose different levels of threat. Many 
> > Americans may be hesitant to embrace such a blatant bias, since it seems to

> > place speculative guilt on many people who are not actually terrorists.
> >
> > It is statistically true that, in the current era, a white elderly woman is

> > less likely to carry a suicide bomb than a young Arabic man. Despite this 
> > statistic, it may be preferable to live in a society in which we act as if 
> > the likelihood of being a terrorist is scattered randomly across all 
> > demographics. If this is the choice we make, then we *all* have to put up 
> > with having our shoes checked for explosives, and having ourselves randomly

> > selected to be patted down, and having to engage in other security measures

> > which cast a broad net. The idea behind banning liquids on planes is not
that 
> > it will deter a highly skilled attacker. The idea is that it will make it 
> > more difficult for a terrorist with average intelligence to cause
unspeakably 
> > horrific damage. Furthermore, the hope is that, in concert with other a 
> > multitude of other mechanisms like X-rays and metal detectors, banning 
> > liquids will help to thwart the vast majority of more sophisticated
attacks. 
> > We all have to put up with these security hassles because, as a society, we

> > don't want airports to subject Muslims or Arabic-looking people to a 
> > disproportionate number of security checks and let everyone else stroll off

> > to their destination gate.
> >
> > With respect to the specific controversy over the liquid ban, airports have

> > an obligation to cover the most obvious threat vectors, particularly those 
> > which the intelligence community has deemed likely to be used in the near 
> > future. Just because we can't detect all threats with perfect accuracy is
no 
> > reason to stop trying to detect them. Judging from Daniel's email, it seems

> > like American airports are being fairly lax about checking whether
passengers 
> > are actually carrying liquids. The same is not true in the UK, where I 
> > currently am. For the past four days in the UK, you haven't been allowed to

> > bring *any* carry-on items onto the plane, with the exception of a
passport, 
> > prescription medication, and baby formula, all of which must be carried in
a 
> > clear plastic bag; any baby formula must be tasted by a parent in front of
a 
> > security agent. All of your luggage must be opened and physically inspected

> > by a security agent, and there is an extremely high probability that your 
> > body will be patted down for any potentially dangerous items, including 
> > liquids. The British aren't kidding around, and their ban on liquids is 
> > serious. If the American airports are not being this intensive, then I can 
> > see how their "ban" on liquids might seem silly. Nevertheless, I think that

> > every bit helps, and when I return to America in two weeks, I'll personally

> > feel safer when I step onto my plane at Heathrow and I know that a relevant

> > threat vector has been extensively checked by the British Airport
Authority.
> >
> > ~j
> >
> 
> -- 
> http://ai.eecs.umich.edu/people/dreeves  - -	search://"Daniel Reeves"
> 
> "Theoretically, there is no difference between theory and practice;
> In practice, there is."
> 
>