Detection and localization of change points in temporal networks with the aid of stochastic block models

Ridder, S. De; Vandermarliere, Benjamin; Ryckebusch, Jan

doi:10.1088/1742-5468/2016/11/113302

Cited by 25 publications

(11 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Existing results on change-point detection for the case when the data sequence takes values in a general metric space are quite limited. Parametric approaches exist for change detection in a sequence of networks (De Ridder, Vandermarliere and Ryckebusch, 2016;Peel and Clauset, 2015;Wang, Yu and Rinaldo, 2018). These approaches have been developed for special cases and are not applicable more generally.…”

mentioning

confidence: 99%

Fréchet Change Point Detection

Dubey¹,

Müller²

2019

Preprint

View full text Add to dashboard Cite

We propose a method to infer the presence and location of changepoints in the distribution of a sequence of independent data taking values in a general metric space, where change-points are viewed as locations at which the distribution of the data sequence changes abruptly in terms of either its Fréchet mean or Fréchet variance or both. The proposed method is based on comparisons of Fréchet variances before and after putative change-point locations and does not require a tuning parameter except for the specification of cut-off intervals near the endpoints where change-points are assumed not to occur. Our results include theoretical guarantees for consistency of the test under contiguous alternatives when a change-point exists and also for consistency of the estimated location of the change-point if it exists, where under the null hypothesis of no change-point the limit distribution of the proposed scan function is the square of a standardized Brownian Bridge. These consistency results are applicable for a broad class of metric spaces under mild entropy conditions. Examples include the space of univariate probability distributions and the space of graph Laplacians for networks. Simulation studies demonstrate the effectiveness of the proposed methods, both for inferring the presence of a change-point and estimating its location. We also develop theory that justifies bootstrap-based inference and illustrate the new approach with sequences of maternal fertility distributions and communication networks.

show abstract

mentioning

confidence: 99%

Fréchet Change Point Detection

Dubey¹,

Müller²

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…One of most important features of the proposed method is detecting changes in meso-scopic network properties, grouping of nodes, such as homophilous or heterophilous groups and core-periphery substructures in a network. The emergence (and changes) of group structures is commonly observed in realworld network data (Borgatti and Everett, 1999;Nowicki and Snijders, 2001;Newman, 2006;Fortunato, 2010;Xu, 2015;Ridder et al, 2016). The formulation of MTRM in Equation (4), however, is designed to recover consistent regression parameters (β) considering network effects as a nuisance parameter.…”

Section: Degree Correctionmentioning

confidence: 99%

“…The motivation of our method is firmly based on a common observation in network analysis that longitudinal network datasets frequently exhibit irregular dynamics, implying multiple changes in their data generating processes (e.g. Guo et al, 2007;Heard et al, 2010;Wang et al, 2014;Cribben and Yu, 2016;Barnett and Onnela, 2016;Ridder et al, 2016). Figure 1 shows an example of longitudinal network data with 3 layers of time series and 90 nodes.…”

Section: Introductionmentioning

confidence: 99%

“…For instance, the maximum size of network analyzed was 11 nodes in Guo et al (2007) and 6 nodes in Wang et al (2014). 3 By incorporating the stochastic blockmodel (SBM) framework, which presumes the existence of discrete node groups, Ridder et al (2016) propose a method to identify a single parametric break. Ridder et al (2016)'s method compares the bootstrapped distribution of the log-likelihood ratio between a null model and an alternative.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Detecting Structural Changes in Longitudinal Network Data

Park¹,

Sohn²

2020

Bayesian Anal.

View full text Add to dashboard Cite

Dynamic modeling of longitudinal networks has been an increasingly important topic in applied research. While longitudinal network data commonly exhibit dramatic changes in its structures, existing methods have largely focused on modeling smooth topological changes over time. In this paper, we develop a hidden Markov multilinear tensor model (HMTM) that combines the multilinear tensor regression model (Hoff, 2011) with a hidden Markov model using Bayesian inference. We model changes in network structure as shifts in discrete states yielding particular sets of network generating parameters. Our simulation results demonstrate that the proposed method correctly detects the number, locations, and types of changes in latent node characteristics. We apply the proposed method to international military alliance networks to find structural changes in the coalition structure among nations. arXiv:1811.01384v1 [stat.ME] 4 Nov 2018 1. group-structured networks with no break (Table 1)

show abstract

“…Most approaches, however, rely on a characteristic time scale on which they describe the dynamics. These can be divided, roughly, into approaches that model temporal correlations via Markov chains relating short-time memory with future behavior [12,13], and those that model the dynamics at longer times, usually via network snapshots [14][15][16][17][18][19] or discrete change points [20][21][22]. In reality, however, most systems exhibit both kinds of dynamics, and focusing on a single aspect comes at the expense of ignoring the other.…”

Section: Introductionmentioning

confidence: 99%

Change points, memory and epidemic spreading in temporal networks

Peixoto¹,

Gauvin²

2017

Preprint

View full text Add to dashboard Cite

Dynamic networks exhibit temporal patterns that vary across different time scales, all of which can potentially affect processes that take place on the network. However, most data-driven approaches used to model timevarying networks attempt to capture only a single characteristic time scale in isolation -typically associated with the short-time memory of a Markov chain or with long-time abrupt changes caused by external or systemic events. Here we propose a unified approach to model both aspects simultaneously, detecting short and longtime behaviors of temporal networks. We do so by developing an arbitrary-order mixed Markov model with change points, and using a nonparametric Bayesian formulation that allows the Markov order and the position of change points to be determined from data without overfitting. In addition, we evaluate the quality of the multiscale model in its capacity to reproduce the spreading of epidemics on the temporal network, and we show that describing multiple time scales simultaneously has a synergistic effect, where statistically significant features are uncovered that otherwise would remain hidden by treating each time scale independently.

show abstract

Detection and localization of change points in temporal networks with the aid of stochastic block models

Cited by 25 publications

References 25 publications

Fréchet Change Point Detection

Fréchet Change Point Detection

Detecting Structural Changes in Longitudinal Network Data

Change points, memory and epidemic spreading in temporal networks

Contact Info

Product

Resources

About