登普斯特-沙弗理论
信念结构
可靠性
计算机科学
熵(时间箭头)
扩展(谓词逻辑)
人工智能
规范化(社会学)
数据挖掘
机器学习
政治学
社会学
物理
程序设计语言
法学
量子力学
人类学
标识
DOI:10.1016/0020-0255(87)90007-7
摘要
We discuss the basic concepts of the Dempster-Shafer approach, basic probability assignments, belief functions, and probability functions. We discuss how to represent various types of knowledge in this framework. We discuss measures of entropy and specificity for belief structures. We discuss the combination and extension of belief structures. We introduce some concerns associated with the Dempster rule of combination inherent in the normalization due to conflict. We introduce two alternative techniques for combining belief structures. The first uses Dempster's rule, while the second is based upon a modification of this rule. We discuss the issue of credibility of a witness.
科研通智能强力驱动
Strongly Powered by AbleSci AI