Mining the K-Most Interesting Frequent Patterns Sequentially

摘要: Conventional frequent pattern mining algorithms require users to specify some minimum support threshold, which is not easy identify without knowledge about the datasets in advance. This difficulty leads dilemma that either they may lose useful information or be able screen for interesting from huge presented patterns sets. Mining top-k allows control number of discovered analyzing. In this paper, we propose an optimized version ExMiner, called OExMiner, mine a large scale dataset efficiently and effectively. order improve user-friendliness also performance system proposed other 2 methods, extended Seq-Miner Seq-BOMA sequentially. Experiments on both synthetic real data show our methods are much more efficient effective compared existing ones.