Study On Mining Algorithm Of Target Frequent Itemsets And Appliction

With the development of information technology, especially the emerging of the network technology, our abilities to collect, store and transfer data have been improved dramatically Comparing to the explosive growth of data, the needs for decision-relevant knowledge are not satisfied yet. Knowledge discovery and data mining technology is an important approach to address this problem.As one of the main patterns in the field of data mining, association rules are used to determine the relationships among the attributes or objects, to find out valuable dependencies among the fields. The efficiency of mining frequent item sets is the key problem in association rules generating. Frequent item sets can be divided into three types: complete, closed and maximal. This dissertation studied thoroughly the related defines, the mining methods of complete and maximal frequent item sets.Target frequent item sets mining algorithms are a sort of that satisfied the users. TFP-tree can filter the items and transactions which don't contribute to target pattern, and compress the integrity algorithms and no-redundant data of the database into a tree, so it can greatly reduce the hunting range. And the algorithms base on SFP-tree are high performances for mining frequent patterns. In the second and the third of the paper, we study on the construct of TFP-tree and STP-tree and the frequent item sets mining algorithms and maximal algorithms frequent item sets mining base on them, then combining both advantage of them, we presented a sorted and compressed and non-redundant target frequent pattern tree (abbreviated to STFP-tree). And base on the STFP-tree, we put forward a target frequent item sets mining algorithm, and a maximal target frequent item sets mining algorithm. The experimental results show that the algorithms are efficiency.With the development and popularization of the Internet, the problems of network safety become severe increasingly. Intrusion detection is a safe strategy for fetching up the shortage of fireproofing. The, fourth of the paper had a try studying on the application of association rules on intrusion detection , it expanded the algorithm STFP-growth and put it into the IDS (Intrusion Detection System) , then evaluated the model through the detection of KDD CUP 99 data.