Abstract

The analysis of consumer-related and consumer-generated data for measuring the success of online retailing is gaining increasing importance. Software packages for data analysis have become commonplace. However, two major shortcomings exist. First, most software solutions are not offered as a service reachable by standard procedures over the Internet, but as isolated stand-alone applications or ERP system modules. Second, privacy restrictions need to be integrated into a framework of business analytics for Web retailers. Whereas the first aspect can be addressed with standardized developer software for Web services, the second must consider privacy legislation, privacy specifications on Web sites (P3P), and data re-identification problems. To address these shortcomings, we propose a formal model of these problems and an implementation of this model as a declarative specification of privacy constraints, expressed as an extension of P3P. The constraints are complemented by a logic that identifies, in a given set of Web analytics, those that might lead to data re-identification and would therefore violate implicit privacy constraints. We present a Webbased service that uses these components to automatically adapt the set of available Web analytics to an online retailer’s P3P policy. The system was tested on a large data set from a major European multi-channel retailer.

Citations