Detection of associations between datasets

A method, program, and system wherein one or more processors identifies a dataset, a set of output class determinations made for entries of the dataset by a computer decisions algorithm, and an undesired disparity between output class determinations resulting from a first value of a first attribute of the dataset and output class determination resulting from a second value of the first attribute. A value of a second attribute is determined to contribute to the undesired disparity through provision to an association rule mining model a first group of the data entries having the first value of the first attribute and a second group of the data entries having the second value of the first attribute, and selecting the value of the second attribute from a set of candidate attributes and values produced by the association rule mining model based on a lift calculation. A request may be received from a user to identify values of one or more attributes other than the first attribute that are contributing to the disparity and a response may be made upon determination that the value of the attribute is contributing to the disparity.