时间序列表示是时间序列挖掘的一个基础和关键问题。对当前出现的各种典型的时间序列表示方法进行了综述,对各自的特点从多个角度进行了比较研究。结果说明,大部分时间序列表示方法将时间序列降维,且都与应用领域紧密相关,在实际构建系统时仍需对各种表示方法按照实际需求进
[22] Korn, F., Jagadish, H. and Faloutsos, C. Efficiently supporting ad hoc queries in large
datasets of time sequences. In: Proc. ACM SIGMOD International Conference on Management of Data, Tuescon, AZ. USA, 1997. 289~300.
[23] Refiei, D. On similarity based queries for time series data. In: Proc. 15th IEEE
International Conference on Data Engineering (ICDE), Sydney, Australia, 1999. 410~417.
[24] Yi, B, K. Faloutsos, C. Fast time sequence indexing for arbitrary LP norms. In: Proc. Of
the 26th International Conference on Very Large Databases(VLDB), Cairo, Egypt. 2000.
[25] Zhu Yunyue. High Performance Data Mining in Time Series: Techniques and Case
Studies: [Ph.D. Thesis], New York University. 2004.
[26] 李斌, 谭立湘,章劲松 等.面向数据挖掘的时间序列符号化方法研究.电路与系统学
报.5(2), 2000.9~14.
[27] 李爱国, 覃征.大规模时间序列数据库降维及相似搜索.计算机学报,2005.1147~1475.
[28] 黄超, 朱扬勇. 基于回归系数的时间序列维约简与相似性查找.模式识别与人工智
能.2006.52~57.
[29] 王元珍, 李俊奎, 曹忠升. RPAA:一种基于时间特性的时间序列表示..计算机科
学.(录用待发表)
Progress and Comparative Study of Time Series
Representations: In the Context of Time Series Mining and
Modeling
LI Junkui WANG Yuanzhen LIU Chengcheng CAO Zhongsheng
Research Institute of Database & Multimedia, Huazhong University of Science & Technology,
Wuhan, Hubei, 430074, China
Abstract
The representation of time series is one of basic and fundamental questions in mining time series sequences. The paper gives a detailed survey on the existing methods for representing time series in the process of mining, and conducts a comparative study in different views. The results show that, most of the representing methods try their best to approximate time series into a lower-dimensional style, and they all domain application depended. There should undergo a transformation process of the representations in the real system construction
Keywords: Data mining Time series Representation Modeling
李俊奎:男,1981年生,博士研究生,主要研究方向是数据挖掘、机器学习。
王元珍:女,1945年生,教授,博士生导师,主要研究方向是现代数据库理论与实现技术、数据挖掘中间件技术。
刘城成:男,1982年生,硕士研究生,主要研究方向是时间序列建模, 多媒体数据库技术。 曹忠升:男,1965年生,副教授,主要研究方向是空间和多媒体数据库技术。
- 6 -
百度搜索“77cn”或“免费范文网”即可找到本站免费阅读全部范文。收藏本站方便下次阅读,免费范文网,提供经典小说教育文库时间序列表示进展及比较研究时间序列挖掘建模环境(6)在线全文阅读。
相关推荐: