论文部分内容阅读
针对关联分类规则产生的候选规则过多导致效率不高的问题,提出一种基于频繁闭项集组成的扩展概念格的分类规则获取方法.利用频繁闭项集提出一种新的概念格模型,通过性质和定理对概念格结点进行剪枝,以抽取分类尽量少且最有效的关联分类规则.研究结果表明:该算法能挖掘出高质量且包含重要信息的关联分类规则,并大大减少关联分类规则的数量,在分类准确率上比现有的关联分类典型算法更高.
Aiming at the problem of low efficiency caused by too many candidate rules generated by association rules, this paper proposes a new method of obtaining classification rules based on frequent closed itemsets and extended concept lattice. A new concept lattice model is proposed based on frequent closed itemsets. Pruning concept lattice nodes by the nature and the theorem to extract the least relevant and effective classification rules.The research results show that the algorithm can mine the association rules of high quality and contains important information and greatly reduce the association The number of classification rules is higher than the typical association classification algorithm in classification accuracy.