Carnegie Mellon University researchers have demonstrated that Social Security numbers (SSNs) can be predicted based solely on an individual’s date and location of birth using statistical techniques. They describe their research in the Proceedings of the National Academy of Sciences as “an unexpected consequence of the interaction between multiple data sources, trends in information exposure, and antifraud policy initiatives with unintended effects.” The discovery makes the U.S. Social Security numbering system vulnerable to fraud, with the researchers noting that it is now possible to regularly reconstruct sensitive personal information from the kind of online postings often found on social networking sites and other online sources. The researchers used an algorithm on 500,000 publicly available records in the Social Security Administration’s Death Master File to successfully identify statistical patterns that then allowed extrapolation to the living U.S. population, making it possible to identify millions of SSNs for individuals whose birth date and location were a matter of public record. The researchers’ sample showed that it was possible to identify in a single attempt the first five digits for 44 percent of deceased individuals who were born after 1988 and for 7 percent of those born from 1973 to 1988, while the identification of all nine digits for 8.5 percent of those born after 1988 was possible in less than 1,000 tries. The prediction system’s accuracy rose for smaller states and for individuals born after 1988 on account of rules that led increasingly to the designation of Social Security numbers at birth. Mark Lassiter with the Social Security Administration has downplayed the significance of the researchers’ conclusions, calling their findings “an exaggeration.”
For more information please visit: http://www.cpccci.com
This entry was posted on Thursday, July 9th, 2009 at 12:02 pm and is filed under Business. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.