Evidential Information Fusion on Possibilistic Structure
The article discusses a new framework for evidential information fusion based on a possibilistic structure. This framework addresses limitations of Dempster's rule by introducing a reversible transformation and a belief evolution network. The proposed method offers more flexibility in combining information from diverse sources and managing conflicts.
- ▪Dempster's rule has strong structural restrictions that limit its flexibility in complex information fusion scenarios.
- ▪The authors propose a reversible transformation derived from the isopignistic principle to enhance belief function representation.
- ▪The new framework supports flexible combination behaviors and is advantageous for non-distinct source fusion and heterogeneous information fusion.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.17038 (cs) [Submitted on 16 May 2026] Title:Evidential Information Fusion on Possibilistic Structure Authors:Qianli Zhou, Ye Cui, Zhen Li, Witold Pedrycz, Yong Deng View a PDF of the paper titled Evidential Information Fusion on Possibilistic Structure, by Qianli Zhou and 4 other authors View PDF HTML (experimental) Abstract:Dempster's rule is a fundamental tool for combining belief functions from distinct and reliable sources. However, its intersection-based semantics imposes strong structural restrictions, which limits its flexibility in handling complex source states and diverse information fusion scenarios.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
