Responsibly innovating data mining and profiling tools: A new approach to discrimination sensitive and privacy sensitive attributes

Abstract

Data mining is a technology that extracts useful information, such as patterns and trends, from large amounts of data. The privacy sensitive input data and the output data that is often used for selections deserve protection against abuse. In this paper we describe one of the main results of our research project on developing new privacy preserving and discrimination aware data mining tools, namely why the common measures for mitigating privacy and discrimination concerns, such as a priori limiting measures (particularly access controls, anonymity and purpose specification) are mechanisms that are increasingly failing solutions against privacy and discrimination issues in the novel context of advanced data mining and profiling. Contrary to previous attempts to protect privacy and prevent discrimination in data mining, we did not focus on new designs that better enable (a priori) access limiting measures regarding input data, but rather focused on (a posteriori) responsibility and transparency. Instead of limiting access to data, which is increasingly hard to enforce in a world of automated and interlinked databases and information networks, rather the question how data can and may be used was stressed.