Multi-dimensional data aggregation utilizing extended partitioned Bonferroni mean Operator

摘要

In this contribution, we develop the concept of an Extended Partitioned Bonferroni Mean $left({{mathcal{E}}{mathcal{P}}{mathcal{B}}{mathcal{M}}}ight)$ operator, which is efficient enough to aggregate input vectors with a varying number of components integrated with some dependence pattern. The global monotonicity for the ${mathcal{E}}{mathcal{P}}{mathcal{B}}{mathcal{M}}$ is analyzed by defining a new partition for each arity. Further to illustrate the applicability and feasibility of the proposed extended aggregation operator, an example based on medical device selection is demonstrated. Finally, we present a way to obtain the weights associated with the corresponding ${mathcal{E}}{mathcal{P}}{mathcal{B}}{mathcal{M}}$ operator employing the Max-Entropy technique.