[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [interesting-people Home]
Subject: [IP] 3+ Million passengers to be misidentified?
Begin forwarded message: From: "Stephen D. Poe" <sdpoe@acm.org> Date: August 14, 2007 7:50:02 PM EDT To: Dave Farber <dave@farber.net> Subject: 3+ Million passengers to be misidentified? Reply-To: sdpoe@acm.org Dave - For IP if you like.Putting together two different sets of facts and running what we used to call a back-of-the-envelope calculation:
"The government proposed a third version of its airline passenger pre- screening program... With the full name, we can resolve 95 percent of the cases correctly. The date of birth adds 3.5 percent to that, and the gender adds another one percent," Hawley said." ('Feds offer simpler flight screening plan', http://news.yahoo.com/s/ap/20070809/ ap_on_go_ca_st_pe/secure_flight)
So I believe Hawley is stating the final error rate is 0.5% mis- identifications.
The Air Transport Association lists 671.7 million US domestic airline passengers for 2006 ( ATA 2006 Annual Report, http://www.airlines.org/ NR/rdonlyres/0E9E7072- ECC6-4CED-8B8E-6857256935E7/0/2007AnnualReport.pdf).
Using the 2006 passenger figures, that would mean the third version of the system using all three proposed pieces of information would 'only' misidentify 3.36 million passengers a year.
Either Hawley's math is bad or something is wrong. Stephen -------------------------------------------
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [interesting-people Home]
Powered by eList eXpress LLC