Fair pattern discovery.

Data mining is gaining societal momentum due to the ever increasing availability of large amounts of human data, easily collected by a variety of sensing technologies. We are assisting to unprecedented opportunities of understanding human and society behavior that unfortunately is darkened by several risks for human rights: one of this is the unfair discrimination based on the extracted patterns and profiles. Consider the case when a set of patterns extracted from the personal data of a population of individual persons is released for subsequent use in a decision making process, such as, e.g. , granting or denying credit. Decision rules based on such patterns may lead to unfair discrimination, depending on what is represented in the training cases. In this context, we address the discrimination risks resulting from publishing frequent patterns. We present a set of pattern sanitization methods, one for each discrimination measure used in the legal literature, for fair (discrimination-protected) publishing of frequent pattern mining results. Our proposed pattern sanitization methods yield discrimination-protected patterns, while introducing reasonable (controlled) pattern distortion. Finally, the effectiveness of our proposals is assessed by extensive experiments.