Decision-Making under Open World with Generalized Evidence Theory

As an extension of Dempster-Shafer evidence theory, Generalized evidence theory can be used in an open world where some targets are unknown. How to make a proper decision under open world is an open issue. In this paper, we propose a method under uncertain environments based on generalized evidence theory, which is used for decision making. We use K-means clustering and elbow methods based on generalized evidence theory to generate a complete discernment framework. And under the framework, a new decision-making method is proposed. After experimental verification, the newly proposed method can well solve the decision-making problem under open world.