| 一种基于DTW的符号化时间序列聚类算法 |
| |
| 引用本文： | 李迎.一种基于DTW的符号化时间序列聚类算法[J].微型机与应用,2011,30(18):3-5. |
| |
| 作者姓名： | 李迎 |
| |
| 作者单位： | 辽宁师范大学计算机与信息技术学院,辽宁大连,116081 |
| |
| 基金项目： | 国家自然科学基金，辽宁省博士启动基金 |
| |
| 摘 要： | 提出了一种基于DTW的符号化时间序列聚类算法,对降维后得到的不等长符号时间序列进行聚类。该算法首先对时间序列进行降维处理,提取时间序列的关键点,并对其进行符号化;其次利用DTW方法进行相似度计算;最后利用Normal矩阵和FCM方法进行聚类分析。实验结果表明,将DTW方法应用在关键点提取之后的符号化时间序列上,聚类结果的准确率有较好大提高。
|
| 关 键 词： | 时间序列 DTW SAX Normal矩阵 FCM |
Symbolization time series clustering based on DTW |
| |
| Li Ying.Symbolization time series clustering based on DTW[J].Microcomputer & its Applications,2011,30(18):3-5. |
| |
| Authors: | Li Ying |
| |
| Affiliation: | Li Ying(Department of Computer and Science Technology,Liaoning Normal University,Dalian 116081,China) |
| |
| Abstract: | A method of clustering symbolization time series based on DTW is proposed to cluster the unequal dimensional time series obtained by reduction. The key points of the time series are firstly extracted and symbolized. Then the similarity between the two time series is calculated by DTW method. Lastly, the normal matrix and FCM algorithm are employed to cluster the time series. The experimental results show that the accuracy of cluster result obtained by the proposed method is good. |
| |
| Keywords: | time series DTW SAX normal matrix FCM |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
