Visual data mining for higher-level patterns: Discrimination-aware data mining and beyond

An important question facing visualization methods is how to be both general and support open-ended exploratory analysis. In this paper, we propose a visualization approach that can on the one hand be applied to any (classification or association) rules, but that is suited to bringing out features of mined patterns that are especially important in discrimination-aware and privacyaware data mining. We define new interestingness measures for items and rules and show various ways in which these can help in highlighting information in interactive settings. We conclude by arguing how this approach can lead to a new generation of feedback and awareness tools.

Publication year:2011