A requirements taxonomy for reducing Web site privacy vulnerabilities

The increasing use of personal information on Web-based applications can result in unexpected disclosures. Consumers often have only the stated Web site policies as a guide to how their information is used, and thus on which to base their browsing and transaction decisions. However, each policy is different, and it is difficult-if not impossible-for the average user to compare and comprehend these policies. This paper presents a taxonomy of privacy requirements for Web sites. Using goal-mining, the extraction of pre-requirements goals from post-requirements text artefacts, we analysed an initial set of Internet privacy policies to develop. the taxonomy. This taxonomy was then validated during a second goal extraction exercise, involving privacy policies from a range of health care related Web sites. This validation effort enabled further refinement to the taxonomy, culminating in two classes of privacy requirements: protection goals and vulnerabilities. Protection goals express the desired protection of consumer privacy rights, whereas vulnerabilities describe requirements that potentially threaten consumer privacy. The identified taxonomy categories are useful for analysing implicit internal conflicts within privacy policies, the corresponding Web sites, and their manner of operation. These categories can be used by Web site designers to reduce Web site privacy vulnerabilities and ensure that their stated and actual policies are consistent with each other. The same categories can be used by customers to evaluate and understand policies and their limitations. Additionally, the policies have potential use by third-party evaluators of site policies and conflicts.