基于快速形式概念集构造的分类规则挖掘算法

提出了Fccrm算法．它采用划分的方法降低了形式背景的规模，通过求形式概念中的最大概念以及它们的下覆盖获取到全部的概念节点，避免了概念格中的复杂关系．同时，该算法还采用了概念剪枝的方法，降低了形式概念集的规模．分类规则集则从形式概念集中提取得到．实验结果表明该算法在性能上有了很大的改进，而且通过形式概念集获取的规则集对于样本集是完备的．

An Algorithm of Mining Classification Rules Based on Faster Constructing Formal Concepts

A novel algorithm called Fccrm is proposed, in which the dividing method is used to reduce the scale of formal context and the technology is adopted that gets all the lattice nodes by seeking all biggest concepts and their lower covers to avoid the complicated relations in concept lattice. The pre-pruning strategy is applied in the mining process to reduce the size of the formal concepts further, for getting the classification rules from the formal concepts. The experimental results show that the performance of this algorithm is obviously improved, and the classification rules set obtained from formal concepts is complete for the sample set.