Transparency in Data Mining: From Theory to Practice - Discrimination and Privacy in the Information Society - page 289

17.1 Introduction: Transparency, Technology and Prediction
Can human behavior be predicted? A broad variety of governmental initiatives are
using computerized processes to try. Recent advances in mathematics, artificial
intelligence and computer science might render this futuristic scenario possible.
Vast datasets of personal information, available to commercial and governmental
entities, enhance the ability to engage in these ventures, and the appetite to push
them forward.
Governments have a distinct interest in automated individualized predictions to
foresee unlawful actions. This is especially true when such behavior generates
substantial risks or is difficult to enforce. Data mining applications are the
technological tools which make governmental prediction possible. They are
essential to overcome the vast amounts of personal information at the
government's disposal, and the need to analyze the information in real time. These
computer programs automatically work through vast datasets to uncover trends in
personal data. They then apply the novel trends and patterns revealed to other
individuals and actions, while sorting the latter accordingly. In doing so, they try
to figure out what the individuals' next steps would be - who of us has a higher
chance of being a tax evader, criminal, or even terrorist.
The growing use of predictive practices premised upon the analysis of personal
information and powered by data mining, has generated a flurry of negative
reactions and responses. An overall concern is the lack of transparency these
processes entail. A call for transparency emerges from the public, press and even
from the US legislator. 1 A need for transparency is commonly cited when calling
for changes in these initiatives (TAPAC Report, 2004; Cate, 2008; Solove, 2008).
Although echoed across the policy, legal and academic debate, the nature of
transparency in this context is unclear and calls for a rigorous analysis.
Transparency might pertain to different segments of the data mining and
prediction process. In addition, it flows from different, even competing, rationales,
as well as a variety of legal and philosophical backgrounds. When viewed in
concert, they lead to different, at times contradicting, conclusions and practical
recommendations. This chapter makes initial steps in illuminating the true
meaning of transparency in this specific context and provides tools for further
examining this issue.
This chapter begins by briefly describing and explaining the practices of data
mining, when used to predict future human conduct on the basis of previously
collected personal information (Part 2). Part 3 moves to address the flow of
information generated in the prediction process. In doing so, it introduces a helpful
taxonomy regarding four distinct segments within the prediction process. Each
segment presents unique transparency-related challenges. This part also provides
for initial strategies as to how transparency could be achieved at every juncture.
Part 4 commences a brief theoretical analysis seeking the foundations for
transparency requirements in this context. The analysis addresses transparency as
a tool to enhance government efficiency, facilitate crowdsourcing and promote
1 Federal Agency Data Mining Reporting Act 42 U.S.C. § 2000ee-3(c)(2).