Long-winded online privacy policies that, for all most consumers know, could include permission to sell personal data or requests for the souls of their first born, could soon be stripped down to their bare essentials.
The 42-month, $3.75 million project aims to use crowdsourcing, natural language processing and machine learning to build a program that analyzes pages of privacy policies for consumers. In addition to a research team led by CMU computer science professor Norman Sadeh, the project will feature work from law school researchers at Fordham and Stanford University.
Mr. Sadeh said a major Web browsing company already has expressed interest in using the final product that comes out of the research as an accessory to its products.
Mr. Sadeh said the primary goal is to use crowdsourcing to determine the privacy issues that are most important to consumers, then create programs to examine whether companies' privacy policies address the issues in a meaningful manner.
"This gives, finally, power to customers to make informed decisions regarding what website they feel comfortable tracking them and which ones they don't," said Mr. Sadeh.
In addition to helping consumers, Mr. Sadeh said the project could also help policy makers determine which privacy concerns most affect consumers and to identify which company's privacy policies are compliant with current laws.
He also said the project will help researchers and policy makers follow the ever-changing formats of privacy policies to study exactly how much consumers are being asked to share.
"Companies like Google and Facebook are very innovative and as they innovate they need to continually revise their privacy policies. This can help [researchers] have a historic perspective on how these privacy policies change, what's been added, what's been relaxed, and enable policy makers and civilian stakeholders to act faster to respond to consumer's needs," he said.
Deborah M. Todd: email@example.com or 412-263-1652.