The power of knowledge : ethical, legal, and technological aspects of data mining and group profiling in epidemiology

Abstract

With the rise of information and communication technologies, large amounts of data are being generated and stored in databases. In order to get a better grip on these large amounts of data, serious efforts are being made to discover patterns and relations in the data with the help of new techniques. One of these techniques is data mining, an automated analysis aimed at finding patterns and relations in data. Through data mining, characteristics may be ascribed to individuals and groups, thus yielding personal profiles or group profiles. In the case of medical data, profiles may be used in epidemiology in order to solve or contribute to solving aetiological, diagnostic, prognostic, or therapeutic problems. However, the same information may be used for determining selection criteria, such as for insurances, jobs, or loans. The information may also be stigmatising or disturbing. Whereas the collection and processing of personal data is subjected to data protection legislation in Europe, this is not necessarily the case for group profiles. As a result, people are increasingly being judged as members of groups, for instance, as people with the same postcode, consumers of peanut butter, or DNA carriers. The Power of Knowledge offers an in-depth analysis of the possible moral problems of data mining and group profiling in medical data. After an analysis of the moral problems, legal and technological solutions are critically discussed. Legal solutions may be found in data protection law, public health law, and anti-discrimination law. Technological solutions based on cryptography may involve restricting the coupling of data and limiting the identifiability of data. Taking a multidisciplinary approach, the author shows how ethics, law, and technology can supplement each other when providing solutions for the problems of data mining and group profiling.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 60 citations exist upstream.