Abstract
In this work we combine clustering ensembles and semi-supervised clustering to address the ill-posed nature of clustering. We introduce a hybrid approach that extends our previous work on clustering ensembles to situations where some knowledge from the end user is available, by enforcing constraints during the partitioning process. The experimental results show that our constrained ensemble technique is capable of producing a partition that is as good as, or better, than those computed by other semi-supervised clustering approaches.