| 多元时间序列模式异常研究 |
| |
| 引用本文： | 郭小芳,李锋,叶华.多元时间序列模式异常研究[J].信阳师范学院学报(自然科学版),2012(4):555-559. |
| |
| 作者姓名： | 郭小芳 李锋 叶华 |
| |
| 作者单位： | 江苏科技大学计算机科学与工程学院;江苏科技大学电子信息学院 |
| |
| 基金项目： | 江苏省高校自然科学研究项目(10JKB520006) |
| |
| 摘 要： | 为了提高多元时间序列模式异常检测算法的有效性和合理性,在k-近邻局部异常检测算法的基础上,结合基于主元分析的多元时间序列的降维方法,对多元时间序列模式异常进行检测.实验结果验证了该算法对多元时间序列模式异常检测的准确性和有效性.
|
| 关 键 词： | 多元时间序列 主元分析 k-近邻 模式异常检测 |
Outlier Pattern Research of Multivariate Time Series |
| |
| GUO Xiao-fang a,LI Feng b,YE Hua.Outlier Pattern Research of Multivariate Time Series[J].Journal of Xinyang Teachers College(Natural Science Edition),2012(4):555-559. |
| |
| Authors: | GUO Xiao-fang a LI Feng b YE Hua |
| |
| Institution: | a(a.School of Computer Science and Engineering;b.School of Electronics and Information, Jiangsu University of Science and Technology,Zhenjiang 212003,China) |
| |
| Abstract: | In order to improve the efficiency of outlier model detection algorithm of multivariate time series(MTS),based on the k-nearest neighbor local outlier detection algorithm,the principal component analysis of the multivariate time series method for dimensionality reduction method was used to detect anomalies of multivariate time series model.The experimental results show that the proposed algorithm detects MTS outlier pattern series more accurately and more efficiently |
| |
| Keywords: | multivariate time series principal component analysis k-nearest neighbor pattern outlier detection |
| 本文献已被 CNKI 等数据库收录！ |
