Data mining

Data mining, in the counter-terrorist world, refers to various methods used to extract (or try to extract) from a large amount of data key nuggets that can help to track and neutralize terrorists. Sometimes, the data has been collected by government sources; more often, it is personal data collected for commercial reasons - and the individuals giving up that information were unaware that it might be passed on to the government.

The Department of Homeland Security (DHS) funded a study of this practice that has just been published. The study was conducted by a privacy and terrorism committee created in 2005 by the National Research Council. Its findings, according to the press report: "The government should not be building predictive data mining systems that attempt to figure out who among millions is a terrorist...The commission found that the technology would not work and that the inevitable mistakes would be un-American."

Those strong statements made me curious to read the report itself. Unfortunately, it looks as if it's not available on the internet - unless you pay for a hard copy. For something like this, I think at least the executive summary should be widely available. (Thanks to Stefaan.)