基于trie merging机制数据流滑动窗口模型的频繁树模式挖掘.

Due to the good expressing power of the tree-structured data format, the data stream transmitted on the Internet is preferred stored in a tree structure. However, knowledge embedded in a data stream is more likely to be changed over time, considering the problem of finding recent frequent tree patterns over a data stream sliding window, this paper proposed a method SWMiner which used to mine recent frequent tree patterns over sliding window, it used to mined the recent frequent tree pattern at any time step. SWMiner used a trie-based structure to compactly store the tree patterns and trie merging approach to support updating support of pattern efficiently. It conducted extensive experiments on several synthetic and real datasets. The results of experiment show that SWMiner performs much better than the well-known existing algorithm and can find the recent frequent tree pattern efficiently. [ABSTRACT FROM AUTHOR]