並列摘要
This paper designs a novel method for data mining on XML streams, which consists of three phases. Each XML document is first transformed into a sequence. After that, a compact tree structure is built to compress the huge amount of subsequences (subtrees) and keep their frequencies. Finally, an efficient algorithm for mining all frequent patterns is provided. Moreover, it can also be applied for mining maximal patterns. The experimental results show that the proposed method performs well in two different kinds of datasets. Its efficiency on sparse data is better, while the accuracy of its returned results on dense data is better when a few errors are tolerable.
延伸閱讀
- Teng, W. G. (2004). 資料串流環境中之頻繁時間樣式探勘 [doctoral dissertation, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU.2004.02412
- 葉介山、林弘祥（2008）。運用群組化節點建構多版本XML文件管理之研究。Electronic Commerce Studies，6(1)，57-81。https://doi.org/10.29767/ECS.200803.0004
- 邵茗煜（2008）。以類型化內容編輯實現XML結構於數位出版之研究〔碩士論文，國立臺中科技大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0061-1811200915323699
- 陳宇威（2017）。巨量XML真實程度模型之建構與介面設計〔碩士論文，國立臺中科技大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0061-3107201715191100
- Jaysawal, B. P. (2013). Progressive Sequential Pattern Mining across Multiple Data Streams [master's thesis, Yuan Ze University]. Airiti Library. https://doi.org/10.6838/YZU.2013.00043