A new approach for generation of generalized basic probability assignment in the evidence theory

The process of information fusion needs to deal with a large number uncertain multi-source, heterogeneity, inaccuracy, unreliability, and incompleteness. In practical engineering applications, Dempster–Shafer evidence theory is widely used in multi-source owing its effectiveness data fusion. Information sources have an important impact on environment the characteristics complex, unstable, uncertain, incomplete. To address problem, this paper considers situation modeling from closed-world open-world assumption studies generation basic probability assignment incomplete information. A new method proposed generate generalized (GBPA) based triangular fuzzy model under assumption. First, maximum, minimum, mean values for membership function each attribute classification problem can be obtained construct representation model. Then, by calculating length intersection points between sample model, GBPA set empty determined. not only different complex environments simply flexibly, but also less loss processing. Finally, series comprehensive experiments basing UCI sets verify rationality superiority method.