Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

In graph theory, complex structures are studied, as well as the dynamics of the connectivity strength of this structure. However, in the estimation procedure, particular characteristics need to be considered at some level as informative when estimating the characteristics of a group. This work proposes a model that provides dynamic estimation of the network structure based on a model that makes it possible to incorporate hierarchy (individual information) in the process. In addition, we show the feasibility of modeling a complex structure by levels, exemplifying this by cluster analysis as a visualization of the embedding projection reduction space. Our case study is a neuroscience experiment, which needs to estimate the brain connectivity map, that is, to study the information flow of the brain in resting-stage subjects. Methods for estimating group networks can be grouped into the following 4 categories: group-structure (GS), virtual-typical-subject (VTS), common-structure (CS), and individual-structure (IS). These four group-structure estimation methods were compared in the context of the Multiregression Dynamic Models. Results showed that the proposed Bayesian Network Structure Dynamic estimation, using GS and hierarchical dynamic models, accommodates the latent/personal information in the estimation process by extracting the pattern shared between them. Moreover, the cluster analysis estimation corroborates the empirical results and expert judgments.

Original publication

DOI

10.1016/j.patcog.2024.110687

Type

Journal article

Journal

Pattern Recognition

Publication Date

01/11/2024

Volume

155