基于属性的相对约简格快速渐进式构造算法

相对约简格作为简化的概念格,在数据挖掘和知识发现等领域具有广泛的应用.相对约简格的构造在其应用过程中是一个主要问题.本文提出了采用树结构对相对约简格节点进行组织,研究了基于属性的相对约简格渐进式构造算法.相对约简格节点的树结构组织可以约束更新格节点、产生子格节点及新生格节点的子结点的搜索范围,从而可以有效地减少算法的执行时间.该算法不仅为相对约简格的构造提供了一种方法,还解决了在已构造好相对约简格的前提下,增加属性所带来的更新问题.在随机生成的数据集上进行的实验测试表明,本算法的时间性能更优越.

Attribute-based Fast Incremental Algorithm for Building Relative Reduced Concept Lattice

Relative reduced concept lattice, a simplified concept lattice, can be used widely in data mining and knowledge discovery, etc. The main difficulty with relative reduced concept lattice-based system comes from the lattice construction itself. In this paper, tree structure is employed to organize the set of concepts in relative reduced concept lattice. Based on attribute, a fast incremental algorithm is developed. The organization of concepts in tree structure can reduce the search space of update concept no...