中文摘要：

鉴于相关性分析在时间序列数据挖掘领域中的重要性,提出一种基于动态时间弯曲的时间序列异步相关性分析方法。利用动态时间弯曲算法在时间序列数据中获得最优弯曲路径,将路径元素扩展为反映原始时间序列异步相关性的新序列,通过计算新序列之间相关系数来实现原始时间序列的异步相关性度量,进而克服传统方法局限于同步相关性分析的问题。数值实验表明,该方法拓展了时间序列数据的相关性分析研究,具有较强的鲁棒性。

英文摘要：

In views of the importance of correlation analysis in the field of time series,this paper proposed a method of asynchronous correlation analysis based on dynamic time warping. It used dynamic time warping to obtain the best warping path between time series and extended its elements to two new series which reflected the asynchronous correlation between the original time series. Finally,it computed the correlation coefficient between the new series to achieve the measurement of the asynchronous correlation between the original time series. It overcame the problem of traditional method confined to the synchronous correlation. The experimental results demonstrate that the proposed method expands the research of correlation analysis for time series and has a strong robustness.