Abstract: Support is an important objective scale to evaluate rules in data mining, and setting mini-support raises the effect of mining. Whereas, the itemss existing and occuring frequency are quite different, so keep on one mini-support in data mining process is unreasonable. In this paper, a new algorithm of mining multi-mini-support rules is presented, and the results prove that it is effective and reasonable with application in mushroom database.

摘要： 支持度是数据挖掘中度量客观兴趣度的重要指标，最小支持度的设置提高了数据挖掘过程的有效性。但是，由于事件在现实中发生和存在频度上有很大的不一致性，始终保持单一的最小支持度显然是不合理的。该文提出了一种多最小支持度规则的挖掘算法，并应用于蘑菇数据库的挖掘过程，结果证明该算法是合理有效的。

关键词: 知识发现, 支持度, 规则

CLC Number: 

• TP311.52