Embodiments of the present disclosure set forth methods for selecting a preferred data set from distributed databases including a first database having a first probability attribute and a second database having a second probability attribute. One example method may include receiving a first data set from the first database and a second data set from the second database. The first data set includes a first monotonic attribute. The second data set includes a second monotonic attribute. It may further include selecting a candidate data set from one of the first data set and the second data set based on a comparison of the first monotonic attribute and the second monotonic attribute and determining whether the selected candidate data set is the preferred data set.