一种面向船舶制造的数据关联规则挖掘算法

A data association rule mining algorithm for shipbuilding

针对船舶制造海量数据关联规则挖掘过程中,事务集占用空间过多导致挖掘效率较低的问题,提出一种基于局部敏感位图存储结构(locally sensitive hash bitmap,LBM)的LBM-Eclat算法。该算法结合了局部敏感哈希和位图2种数据结构,并可以根据存储数据量的变化动态调整内部数据存储结构。通过对比实验证明基于LBM的LBM-Eclat算法能够有效提升对密集型数据集的挖掘效率,同时减少挖掘过程中的空间消耗。

In order to solve the problem of low mining efficiency caused by too much space occupied by transaction sets in the process of mining association rules of shipbuilding massive data,the Eclat algorithm is used to convert the merging of transactions into set operations using vertical databases.A LBM-Eclat algorithm based on locally sensitive hash bitmap(LBM)is proposed.LBM Eclat combines two data structures,local sensitive hash and bitmap,and can dynamically adjust the internal data storage structure according to the changes of the amount of stored data.Through comparative experiments,it is proved that LBM-Eclat algorithm based on LBM can effectively improve the mining efficiency of dense data sets and reduce the space consumption in the mining process.