因果关系推断是因果顺序理论分析的重要研究内容之一。本文利用属性拓扑理论,提出基于属性拓扑的因果关系可视化推断方法。该方法利用属性拓扑的可视化特点,以属性拓扑中的伴生关系为基础,对属性对所属对象集之间的依赖关系进行分析和推理。通过属性拓扑中属性去除不断地更新形式背景,从而推断出各属性间的因果关系。该算法使因果关系计算精准、可视化且易于实现。%Causal inference is one of the important research contents in the analysis of causal order theory. In this paper, a visual inference method of causal relation is proposed based on the theory of attribute topology. This method makes use of the visual characteristics of attribute topology, based on the companion relation in attribute topology, ana-lyzes and reasoning the dependency between attributes on the set of objects. The formal context is updated by attribute removal in attribute topology, and the causal relationship between attributes is inferred. This algorithm makes the causal relation calculation accurate, visual and easy to implement.
展开▼