Geosocial Graph-Based Community Detection

TL;DR

This paper applies spectral clustering and multislice modularity optimization to LAPD data to detect social communities, comparing results with known gangs and discussing data sparsity and sociological complexities.

Contribution

It introduces a combined spectral and modularity approach for community detection in geosocial data, highlighting challenges in aligning algorithmic communities with real-world groups.

Findings

Detected communities often differ from known gangs due to data sparsity.

Social connections in data are incomplete, affecting community detection accuracy.

Complex sociological factors influence the clarity of community boundaries.

Abstract

We apply spectral clustering and multislice modularity optimization to a Los Angeles Police Department field interview card data set. To detect communities (i.e., cohesive groups of vertices), we use both geographic and social information about stops involving street gang members in the LAPD district of Hollenbeck. We then compare the algorithmically detected communities with known gang identifications and argue that discrepancies are due to sparsity of social connections in the data as well as complex underlying sociological factors that blur distinctions between communities.