Time series pattern identification by hierarchical community detection

Identifying time series patterns is of great importance for many real-world problems in a variety of scientific fields. Here, we present a method to identify time series patterns in multiscale levels based on the hierarchical community representation in a complex network. The construction method transforms the time series into a network according to its segments’ correlation. The constructed network’s quality is evaluated in terms of the largest correlation threshold that reaches the largest main component’s size. The presence of repeated hierarchical patterns is then captured through network metrics, such as the modularity along the community detection process. We show the benefits of the proposed method by testing in one artificial dataset and two real-world time series applications. The results indicate that the method can successfully identify the original data’s hierarchical (micro and macro) characteristics.

[1]  Jun Wang,et al.  A novel exopolysaccharide-producing and long-chain n-alkane degrading bacterium Bacillus licheniformis strain DM-1 with potential application for in-situ enhanced oil recovery , 2020, Scientific Reports.

[2]  Yue Yang,et al.  Complex network-based time series analysis , 2008 .

[3]  M Small,et al.  Complex network from pseudoperiodic time series: topology versus dynamics. , 2006, Physical review letters.

[4]  Hua Xu,et al.  A new time series prediction method based on complex network theory , 2017, 2017 IEEE International Conference on Big Data (Big Data).

[5]  Lucas Lacasa,et al.  From time series to complex networks: The visibility graph , 2008, Proceedings of the National Academy of Sciences.

[6]  Zhong-Ke Gao,et al.  Multivariate weighted complex network analysis for characterizing nonlinear dynamic behavior in two-phase flow , 2015 .

[7]  M. Newman,et al.  Finding community structure in very large networks. , 2004, Physical review. E, Statistical, nonlinear, and soft matter physics.

[8]  Zhong-Ke Gao,et al.  Multiscale complex network for analyzing experimental multivariate time series , 2015 .

[9]  M. Mildner,et al.  Re-epithelialization and immune cell behaviour in an ex vivo human skin model , 2020, Scientific Reports.

[10]  E. Macau,et al.  Spatiotemporal data analysis with chronological networks , 2020, Nature Communications.

[11]  Xiang Li,et al.  Recent advances in bridging time series and complex networks , 2013, 2013 IEEE International Symposium on Circuits and Systems (ISCAS2013).

[12]  Huajiao Li,et al.  Characteristics of the transmission of autoregressive sub-patterns in financial time series , 2014, Scientific Reports.

[13]  Michael Small,et al.  Complex network analysis of time series , 2016 .

[14]  Huijie Yang,et al.  Visibility Graph Based Time Series Analysis , 2015, PloS one.

[15]  David M Blei,et al.  Efficient discovery of overlapping communities in massive networks , 2013, Proceedings of the National Academy of Sciences.

[16]  M E J Newman,et al.  Finding and evaluating community structure in networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[17]  Michael Small,et al.  Complex networks from time series: Capturing dynamics , 2013, 2013 IEEE International Symposium on Circuits and Systems (ISCAS2013).

[18]  Santo Fortunato,et al.  Community detection in graphs , 2009, ArXiv.

[19]  Nicholas A. C. Marino,et al.  Extreme rainfall events alter the trophic structure in bromeliad tanks across the Neotropics , 2020, Nature Communications.

[20]  Leandro Anghinoni,et al.  Temporal Network Pattern Identification by Community Modelling , 2020, Scientific Reports.

[21]  Lucas Lacasa,et al.  Network structure of multivariate time series , 2014, Scientific Reports.

[22]  Durga Toshniwal,et al.  Trend Analysis of Time Series Data Using Data Mining Techniques , 2014, 2014 IEEE International Congress on Big Data.

[23]  Liang Zhao,et al.  Time series clustering via community detection in networks , 2015, Inf. Sci..

[24]  Zhong-Ke Gao,et al.  Multi-frequency complex network from time series for uncovering oil-water flow structure , 2015, Scientific Reports.

[25]  T. Vicsek,et al.  Uncovering the overlapping community structure of complex networks in nature and society , 2005, Nature.

[26]  Jing He,et al.  Cautionary tales on air-quality improvement in Beijing , 2017, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[27]  Dong-Hong Ji,et al.  Time series trend detection and forecasting using complex network topology analysis. , 2019, Neural networks : the official journal of the International Neural Network Society.

[28]  Albert-Lszl Barabsi,et al.  Network Science , 2016, Encyclopedia of Big Data.

[29]  Michael Small,et al.  Recurrence-based time series analysis by means of complex network methods , 2010, Int. J. Bifurc. Chaos.

[30]  Jürgen Kurths,et al.  Recurrence networks—a novel paradigm for nonlinear time series analysis , 2009, 0908.3447.

[31]  Jürgen Kurths,et al.  Ambiguities in recurrence-based complex network representations of time series. , 2010, Physical review. E, Statistical, nonlinear, and soft matter physics.

[32]  Liang Zhao,et al.  Detecting Time Series Periodicity Using Complex Networks , 2014, 2014 Brazilian Conference on Intelligent Systems.