Data Mining as Search: Theoretical Insights and Policy Responses - Discrimination and Privacy in the Information Society - page 320

intrusion is a complicated notion. Some individuals might be greatly troubled by
the automated nature of the data mining process, and the lack of human decision-
making and discretion. Yet others might have a very different set of preferences
when it comes to governmental analyses of personal data. To properly assess the
notion of psychological intrusion at this specific juncture, one must remember the
alternative strategy to governmental data mining. This would call for broader roles
for experts and field officers in the law enforcement decision making process; in
such a non-automated process, actual humans are those sifting and considering
the individual's files. For some individuals, data mining generates greater
anxiety than this latter options given concerns with automated and computerized
decision-making processes. For others, however, the opposite would be true. 3
These persons would not be alarmed by the faceless computer searching their data
(Tokson, 2011). They would, however, be gravely concerned with actual indivi-
duals looking through their information.
A similar complication will follow when considering the psychological
intrusion resulting from fears of powerful revelations made by a computer
algorithm. While this is the perspective of some, others might have greater fears of
the other practices government might apply if data mining is set aside. When
relying on experts and field officers, the process might be ridden with errors and
biases which result from the cognitive limitations and opinions of humans
(Zarsky, 2012). These are concerns that the computer analysis could avoid with
greater success.
The last few paragraphs set out arguments which explain that data mining
processes might generate a sense of psychological intrusion for some, yet might be
comforting to others. The latter are individuals whom believe that this process is
preferable to its inevitable alternatives. Both arguments and points of view seem
acceptable, even reasonable. The differences of opinion people will have regarding
the intrusiveness of data mining will result from differences in their understanding
of the data mining technology, its benefits, and its detriments. A possible measure
to overcome the analytical obstacle this theory faces might be through conducting
surveys to establish the public's position. Yet administering such surveys would be
a very difficult, perhaps near-impossible task (Solove, 2010).
To conclude, the “psychological intrusion” perspective to the law of searches
can easily be applied to the context of governmental data mining. It can easily
explain why, for some, the governmental actions breach privacy rights. However,
this perspective - if ever accepted and applied by law - will face problems when
moving from theory to practice. Establishing whether data mining is indeed
intrusive will depend on a variety of unpredictable factors. Thus, this theory will
probably fail to provide clear-cut policy.
Limiting Searches to Limit the Force of Government
A second theory distinguishing between legal and illegal searches which can
illuminate the privacy-in-data mining debate looks to the normative reasons (as
3 For instance, see Goldman, Data Mining and Attention Consumption, 225, 228, as
discussed by S OLOVE , N OTHING TO H IDE , supra note 7, at 183.