Abstract:One feature discovered in the study of complex networks is community structure, in which vertices are gathered into several groups where more edges exist within groups than between groups. Many approaches have been developed for identifying communities; these approaches essentially segment networks based on topological structure or the attribute similarity of vertices, while few approaches consider the spatial character of the networks. Many complex networks are spatially constrained such that the vertices and edges are embedded in space. In geographical space, nearer objects are more related than distant objects. Thus, the relations among vertices are defined not only by the links connecting them but also by the distance between them. In this article, we propose a geo-distance-based method of detecting communities in spatially constrained networks to identify communities that are both highly topologically connected and spatially clustered. The algorithm is based on the fast modularity maximisation (CNM) algorithm. First, we modify the modularity to geo-modularity Qgeo by introducing an edge weight that is the inverse of the geographic distance to the power of n. Then, we propose the concept of a spatial clustering coefficient as a measure of clustering of the network to determine the power value n of the distance. The algorithm is tested with China air transport network and BrightKite social network data-sets. The segmentation of the China air transport network is similar to the seven economic regions of China. The segmentation of the BrightKite social network shows the regionality of social groups and identifies the dynamic social groups that reflect users’ location changes. The algorithm is useful in exploring the interaction and clustering properties of geographical phenomena and providing timely location-based services for a group of people.

Density-based place clustering in geo-social networks

Density Peak Clustering with connectivity estimation

Comparative Density Peaks Clustering

Finding community structure in spatially constrained complex networks

A Statistical Information-Based Clustering Approach in Distance Space

DBSTC: an Effective Method for Discovering Cluster Features with Different Spatiotemporal Densities

Discovering Geo-dependent Stories by Combining Density-based Clustering and Thread-based Aggregation techniques

Geographical Impacts on Social Networks from Perspectives of Space and Place: an Empirical Study Using Mobile Phone Data

HiSpatialCluster: A Novel High-Performance Software Tool for Clustering Massive Spatial Points.

Distributed Information Theoretic Clustering

Discovering Spatial Interaction Communities from Mobile Phone Data

Spatially-constrained clustering of ecological networks

An improved density peaks-based clustering method for social circle discovery in social networks

How Urban Places Are Visited by Social Groups? Evidence from Matrix Factorization on Mobile Phone Data

A Geovisual Analytic Approach to Understanding Geo-Social Relationships in the International Trade Network.

A Socio-geographic Perspective on Human Activities in Social Media

Spatial Co-Location Pattern Mining Based on the Improved Density Peak Clustering and the Fuzzy Neighbor Relationship.

A scalable multi-density clustering approach to detect city hotspots in a smart city

On geo-social network services

Density-ratio Based Clustering for Discovering Clusters with Varying Densities.

Density-Based Clustering On Massive Mobile Communication Data