Spatial network disintegration based on kernel density estimation

doi:10.1016/j.ress.2024.110005

Spatial network disintegration based on kernel density estimation

2024-05-01

5

PRE

AI

DOI：10.1016/j.ress.2024.110005

Original

Original request for help

Save

Abstract

En 中文

The problem of network disintegration, such as suppression of an epidemic spread and destabilization of terrorist networks, possesses extensive applications and has lately been the focus of growing interest. Many realworld complex systems are represented by spatial networks in which nodes and edges are spatially embedded. However, existing disintegration approaches for spatial network disintegration focus on singular aspects such as geospatial information or network topography, with insufficient modeling granularity. In this paper, we propose an effective and computationally efficient virtual node model that essentially integrates the geospatial information and topology of the network by modeling edges as virtual nodes with weights. Moreover, we employ Kernel Density Estimation, a well-known non -parametric technique for estimating the underlying probability density function of samples, to fit all nodes, comprising both network and virtual nodes, to identify the critical region of the spatial network, which is also the circular geographic region where disintegration occurs. Extensive numerical experiments on synthetic and real -world networks demonstrate that our method outperforms existing methods in terms of both effectiveness and efficiency, which provides a fresh perspective for modeling spatial networks.

Keywords:

Spatial network

Network disintegration

Disintegration circles

Virtual node model

Kernel density estimation

Spatial network disintegration based on kernel density estimation

Abstract

Journal

Organization