摘要
针对因果图的精确推理是NP难的,提出寻找近似的推理算法。根据以往文献中近似推理的原理,通过一种可能性比值,找到转化后的连接事件概率。该近似推理保证了多值因果图在推理过程中概率的归一性,最后用于实例得出的结果满足概率论知识且符合实际。
The exact inference of causal graph is NP and hard, and the inference algorithm was proposed. Through a possibility ratio, the probability of an event was found after transformation according to the principle of approximate reasoning from previous literature. This approximate reasoning ensured the unification of probability of the multiple value causality diagrams in the reasoning process, and the application to examples drawn from the results meet the knowledge of probability theory and is also in line with the actual.
出处
《重庆理工大学学报（自然科学）》
CAS
2016年第4期127-131,共5页
Journal of Chongqing University of Technology：Natural Science
基金
国家社科基金资助项目(13BTJ008)
关键词
因果图
归一性
近似推理
causality diagram
normalization
approximate reasoning
