XML频繁子树挖掘研究

本文主要研究通过引入滑动窗口模型，提出了面向XML查询数据流的频繁查询子树在线挖掘算法。算法以全局Trie树为数据缓存结构管理和维护缓存池的数据流，采用自下而上、基于前缀等价类的遍历方式快速产生所有带根查询子树和标准查询子树。实验结果表明该算法具有挖掘速度快、内存消耗稳定等特点，可以有效、平稳地处理XML查询数据流。

Research the Mining of Frequent XML

This paper studies the problem that we propose an online mining of frequent XML query patterns is stream. In this problem, we employ the sliding window model to control the memory which stores the sample XML queries. To better maintain the query stream in the pool we introduce a data structure named global trie. And to generate sub-queries, we enumerate the query from bottom to top based on the concept of prefix equivalence class. Experiments show that our approach can deal with XML query stream effectively and stably.