A probabilistic transformation of basic probability assignment (BPA) in D-S evidence theory

The BPA in the Dempster-Shafer (D-S) evidence theory can effectively represent and process the uncertainty information and it has been widely used in various fields. But the problem of how to make decisions based on BPA still needs to be solved. To transform BPA into probability function is a simple and feasible solution, so we propose what we believe to be a new method for transforming BPA into probability by utilizing the information contained in the belief function and plausibility function of the propositions in the D-S evidence theory. Compared with the existing Pignistic Probability Transform based on the transferable belief model and the plausibility function transform, our probability transformation method can more effectively utilize the known information of a target identification system and achieve the reasonable transformation from BPA to probability distribution. A numerical example also verifies that our method can measure the effects of the belief function and the plausibility function on the distribution of BPA to the multi-subset propositions. All these indicate that our method is effective.