形式概念分析是一种强有力的数据分析和可视化工具，自提出至今已被应用到数据挖掘、知识发现等领域。但由于概念间复杂的关系使得概念格中的连线错综复杂，尤其是在处理大的形式背景时显得尤为混乱。形式背景中所包含的属性间关系、对象间关系以及对象-属性关系是形式背景的本质关系，本文在人类认知事物哲学原理的指导下，构建了以描述属性间关系和区分对象为基本目的的偏序结构图，并描述了其构建方法。该种图形层次分明、结构清晰，不存在线的交叉连接，具有比较好的可视化效果，尤其计算方法简单，存在应用到大数据的潜能，为大数据数据分析和知识发现提供了一种潜在工具。%Formal concept analysis is a powerful tool in data analysis and visualization, and has been applied to data mining, knowledge discovery and many other fields since proposed. However, in the concept lattice, the complex relations between concepts make the lines rather complicated and crossed, especially when dealing with a large-scale formal context. The relation among at-tributes, objects and attribute-object are the essential relations in a formal context. Therefore, under the guidance in the philosophical principle of human being's cognition, the partial ordered structure diagram aiming to delineate the relations among attributes and distinguish distinctive objects is proposed, and construction method is described. Its distinct hierarchy, clear structure, uncrossed lines provide a better visualization. Apart from that, simple computational method of it makes a large potential in allusion to big data. Hence, a novel and efficient tool towards data mining and knowledge discovery of big data is provided by this diagram.
展开▼
