Characterizing ordinal network of time series based on complexity-entropy curve
作者：
Highlights：
• The authors propose an ordinal network-based complexity-entropy curve that can be used to the novelty detection of signal dynamics.
• The proposed method show greater discriminating power and better robustness against noise than the conventional global node entropy.
• By means of the proposed curve, delicate differentiation can be achieved among chaotic signals with extremely similar dynamics.
• Two simple network measures, average out-degree and Gini index of out-degree distribution are efficient to discriminate signals.
• The curve detects abnormal dynamical patterns in the stock markets during the 2008 global financial crisis and identifies the dynamical change in Earth’s geomagnetic activity.
摘要
•The authors propose an ordinal network-based complexity-entropy curve that can be used to the novelty detection of signal dynamics.•The proposed method show greater discriminating power and better robustness against noise than the conventional global node entropy.•By means of the proposed curve, delicate differentiation can be achieved among chaotic signals with extremely similar dynamics.•Two simple network measures, average out-degree and Gini index of out-degree distribution are efficient to discriminate signals.•The curve detects abnormal dynamical patterns in the stock markets during the 2008 global financial crisis and identifies the dynamical change in Earth’s geomagnetic activity.
论文关键词：Ordinal network,Signal processing,Symbolic patterns,Tsallis q-entropy,Novelty detection
论文评审过程：Received 15 June 2020, Revised 23 November 2021, Accepted 26 November 2021, Available online 27 November 2021, Version of Record 5 December 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.108464
