Biased population samples pose a prevalent problem in the social sciences. Therefore, we present two novel methods that are based on positive-unlabeled learning to mitigate bias. Both methods leverage ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results