| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
|---|---|---|---|---|---|---|
| 1 | 2 | |||||
| 3 | 4 |
5 | 6 | 7 | 8 | 9 |
| 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| 17 | 18 | 19 | 20 | 21 | 22 | 23 |
| 24 | 25 | 26 | 27 | 28 | 29 | 30 |
| 31 |
« CFP: INSNA 2007, Corfu | Main | X-mas and Social Networking »
14 December 2006
One of our May posts about an article by David Lazer, Frederick Bieber and Charles Brenner in Science discussed using the DNA of relatives to catch guilty kin. These days I stumbled on an article in The Times about a pilot project in London to identify the highest-risk future offenders based on various data sources which reminded my of this entry. Statistically speaking, a long domestic violence record increases the risk of becoming a murderer. Richard Berk, Professor of Criminology and Statistics at UPenn is currently working on a project, funded by the National Science Foundation, on the development and application of statistical learning procedures for data sets in the behavioral, social and economic sciences. Among the key applications are anticipating failures on probation or parole and forecasting crime “hot spots” a week in advance.
See an interview with him in the Phiadelphia Inuirer. If someone can build relations based on my DNA or crime records, may be so can my social relations. David's past quote is very applicable here:
[...] the more general issue around data mining of relational data, individual choice, and privacy [...], so much information is inherently relational. The fact that I have certain characteristics may say something about people that I have various types of connections to. This does create certain conundrums for constructions of privacy that rely on individual choice, since with relational information, what I choose to reveal about myself reveals something about others-- i.e., there are informational externalities. In this day and age, this issue is endemic, and suggests that certain types of decisions about privacy must be necessarily communal and not individual in nature.
Posted by Alexander Schellong at December 14, 2006 12:01 AM