Title :
Gaussian kernel approximate entropy algorithm for analyzing irregularity of time-series
Author :
Xu, Li-Sheng ; Wang, Kuan-Quan ; Wang, Lu
Author_Institution :
Dept. of Comput. Sci. & Eng., Harbin Inst. of Technol., China
Abstract :
Approximate entropy (ApEn) has been widely used to analyze the complexity of time series. However, the inconsistency that ApEn exhibits not only limits its applications but also raises questions about its validity. Addressing this issue, this paper presents a novel Gaussian kernel approximate entropy (GApEn) algorithm. The experimental results demonstrate that GApEn performs better than ApEn in terms of relative consistency, stability and statistical accuracy.
Keywords :
Gaussian processes; computational complexity; entropy; signal processing; statistical analysis; time series; Gaussian kernel approximate entropy; stability; statistical accuracy; time series complexity; time-series irregularity analysis; Algorithm design and analysis; Application software; Computer science; Entropy; Heart rate variability; Kernel; Pulse measurements; Stability; Time measurement; Time series analysis; Approximate entropy; Gaussian kernel; relative consistency;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527935
