Change-point detection with recurrence networks

Change-point detection based on an observed time series has emerged as an important method for detecting changes in dynamics of real-world systems. Recently, recurrence networks have been shown to be useful, which are network representations of recurrences, to analyze underlying dynamics. In this paper, we propose a new method for detecting dynamical changes using recurrence networks. The proposed method extracts a group of time indices that share the same dynamics as a community of the recurrence network. In addition, some numerical simulations are presented to verify the validity of this method.