Abstract: Typical association rules mining studies on associations between items found in transactions in databases.Negative association rules mining considers absent items in databases as well as existing items.A concept of complete negative association rules is proposed.A tree-based algorithm named Free-PNP is presented by which negative frequent patterns can be found from databases and complete negative association can be obtained.Experiments demonstrate its effectiveness and efficiency.

摘要： 传统的关联规则挖掘研究事务中所包含的项与项之间的关联性，而负关联规则挖掘不仅要考虑事务中包含的项，还要考虑事务中不包含的项。给出了完全负关联规则的定义，提出一种基于树的算法Free-PNP，通过此算法挖掘数据库中的负频繁模式，继而得到所要挖掘的完全负关联规则。通过实验验证了算法的有效性。