Abstract: Fuzzy association rules finds association rules based on fuzzy set theory. Frequent items mining is the key problem in data mining. When frequent sets is being searched in Apriori, database needs to be scanned several times, and a large candidate set is checked through pattern matching. The algorithm running efficiency is reduced. Aiming at this problem, this paper proposes Fuzzy Transaction Data-mining Algorithm 2(FTDA2). This algorithm scans database once, and record the transactions, which contribute to the support of frequent set. It compares FTDA2 to other algorithms, and proves its validity by experiment.

Key words: fuzzy set, association rule, transaction, quantitative attribute

摘要： 模糊关联规则在模糊集理论的基础上发现关联规则，频繁项集挖掘是数据挖掘的关键问题。Apriori算法在查找频繁项集时，需要对数据库进行多次扫描，通过模式匹配检查一个很大的候选集合，降低了算法执行效率。针对该问题提出FTDA2算法，该算法对事务数据库进行一次扫描，记录对计算频繁项集支持度有贡献的事务。比较FTDA2算法与其他算法，通过实验证明其有效性。

关键词: 模糊集, 关联规则, 事务, 数值型属性

CLC Number: 

• TP391