多最小支持度的加权关联规则挖掘算法

摘要

Algorithm of mining weighted association rules with multiple minimum supports is proposed to solve the problems that the transactions and data items have not the same importance in datasets. The algorithm allows user to specify multiple minimum supports and give the weights of the transactions to find the useful association rules. The transactions are classified as the item' s minimum support ascending order in the algorithm, and the weighted frequent itemsets in every classification are solved successively. The redundant items are deleted and the same itemsets are cumulated during the mining, and the new algorithm needn't scan the database many times repeatedly,so the efficiency of the mining is improved. The experiment shows that the algorithm is efficient in mining the weighted association rules from transaction dataset.%针对数据集中交易记录和数据项的重要性不同问题,提出了一种多最小支持度的加权关联规则挖掘算法,允许用户设定多个最小支持度,给出交易记录不同的权重,从而发现有价值的关联规则.该算法按项目的最小支持度升序对交易记录进行分类,按类别依次求出每一类别内的加权频繁集.在挖掘过程中由于剔除了冗余项目并对相同项集累加计数,且不需多次重复扫描数据库,从而提高了挖掘效率.实验结果表明,新算法能有效地从数据集中挖掘出加权关联规则.