My blog has moved! Redirecting...

You should be automatically redirected. If not, visit and update your bookmarks.

Data Mining Research - Data mining for terrorism: is it working? (cont'd)

I'm a Data Miner Collection (T-shirts, Mugs & Mousepads)

All benefits are given to a charity association.

Thursday, February 22, 2007

Data mining for terrorism: is it working? (cont'd)

To continue on the post of yesterday, I want to write a few words about Jonas and Harper paper. In this paper, the authors argue that data mining is definitely not useful against terrorism. The first sentence which avoid all confusion about their position is: "Though data mining has many valuable uses, it is not well suited to the terrorist discovery problem". According to the authors, data mining "would waste taxpayer dollars, needlessly infringe on privacy and civil liberties, and misdirect the valuable time and energy of the men and women in the national security community".

I think that we need to distinguish between the usefulness of data mining for terrorism (i.e. can we extract meaningful knowledge from data) and the possibility to use it in everyday life (i.e. is it effective, too expensive, not ethical?). In the paper, authors only argue that data mining for terrorism waste time and money. Although they have good arguments, they are not directly related to technical aspects of data mining.

I think the discussion is open. Since I'm not in the domain of terrorism data, you should perhaps have a look at Jonas' weblog for details.

Sphere: Related Content


Tuve Löfström said...

data mining is perhaps not as effective in this domain as, for instance, information fusion. The reason would be that we would like to detect new patterns online, from many different sources. but the techniques used in information fusion systems are often from the same (or a similar) toolbox as the one used within the data mining community.

Tuve Löfström said...

It should be added that data mining could be used within IF as part of the solution...
i.e. Llinas et al.

Anonymous said...

I highly recommend Bruce Schneier's postings on this issue: Data Mining for Terrorists and Terrorists, Data Mining, and the Base Rate Fallacy

The extended debates which follow in the comments section are also worth reading.

Sandro Saitta said...

Thanks for the comments and the links. As written in my post, I think that the distinction between the questions Is data mining useful for terrorism? and Should we use data mining for terrorism? is not always clear.

Clicky Web Analytics