Background and central aspects of the topic
The automated analysis of continuously growing public and private data sets is a, if not the, fundamental business model of the digital society. In this project »data mining« refers to the process of knowledge discovery in database by applying mathematical and statistical methods and algorithms. The aim is to identify previously unknown new patterns and correlations, derive usable information from these and, if appropriate, develop prognostic models and expert systems which, based on the data, yield decision-making aids for different user groups. In science this is a methodological approach which has been used for decades to handle large volumes of data in, for instance, physics, the biosciences and the geosciences. In all other areas of society, too, data generation and digitisation have been increasing enormously for years, in areas such as health and finance, in the transport and energy sectors, in the surveillance of public spaces and in public administration (e-government). In connection with the growing technical capabilities offered by ever faster processing, the use of these large and varied data sets (»big data«) by industry and government, but also by civil society stakeholders and private persons raises hopes of individually tailored services and improved means of monitoring a range of business processes. New challenges arise in the areas of freedom of information, informational self-determination, (intellectual) property rights and data protection. For years people in the scientific community, industry, civil society and politics have been discussing the associated potentials for innovation and, increasingly, also legal and ethical issues or regulatory options.
The various issues associated with designing the digital economy or society such that it meets society's demands have been a major area of technology assessment (TA) studies for some time and will remain so in future. A key challenge is to analyse the rapid changes in the scientific and technical options and the socio-economic and societal developments such that the results do not appear to be out of date after a short time. To guarantee this, the current debates must be promptly captured and compared with existing and substantively founded knowledge obtained, for example, from previous TA projects. After all, it should be noted that many of the associated issues are not fundamentally new but have primarily attained new quantitative dimensions and are appearing (now and even more so in future) in new application fields.
Objectives and approach
Based on two case studies, the TA project »data mining – social and legal challenges« will analyse and discuss legal, ethical, political and socio-economic issues raised by data mining. One case study will examine the use of data mining methods in medicine and healthcare, which particularly involves the use of personal health data. The second case study is to address the use of data mining methods in performing public-sector tasks (with the exception of healthcare), with particular reference to the use of geodata for observing and monitoring various processes on Earth (e.g. meteorological services, environmental changes, surveillance of public spaces). The two application fields were selected because of their current and future parliamentary relevance. If public services information is used for data mining, the legislator has a particular obligation to ensure that the system designed to capture and use the data complies with the relevant regulations.
The case studies deal with two very different key applications for data mining, illustrating many of the legal, ethical and political issues. The central questions are: Which traditional and which new stakeholder groups can set up and manage which data sets, and who has access to them? Which types of data sets are already being merged and analysed comparatively and for what purposes? Who »owns« the data, who decides on its processing, and who monitors the processes? Which new business models from the digital economy are complementing existing ones? What level of quality, reliability and validity do the results of automated evaluation have in analytical and prognostic terms? What can the data mining results be used for, and by whom? Where are limits and possibilities due to more detailed tracking and personalisation methods?
In addition to the case studies, important international TA studies and public discourse and participation processes on big data/data mining will be evaluated in order to obtain an overview of the debate, activities and assessments in other countries. Following this, a decision will be made in consultation with the parliamentary TA rapporteurs as to whether an in-depth legal analysis of selected issues (e.g. relating to property, copyright and data protection legislation, possibly with a comparison with international law) seems reasonable or whether the aim should be to generate a larger public discourse, e.g. on the handling of health data.
The final report is currently going through the process of approval by the group's TA rapporteur group.
Publications on the topic
Ferdinand, J.-P.; Kind, S.
2018. Büro für Technikfolgen-Abschätzung beim Deutschen Bundestag (TAB). doi:10.5445/IR/1000133904
Kind, S.; Weide, S.
2017. Büro für Technikfolgen-Abschätzung beim Deutschen Bundestag (TAB). doi:10.5445/IR/1000133902