Demarcating geographic regions using community detection in commuting networks with significant self-loops

Journal article

Marriage networks, which represent the matrimonial connections between different families in a given historical and geographical milieu, rarely take into account one aspect of internal family dynamics, namely the existence of intra-family marriages. The inclusion of such marriages, represented in the graph by self-loops, is essential in order to compute more accurate measures of centrality. In this paper, we discuss various procedures for incorporating these links into the analysis, with the requirement that they be compatible with the use of already available social network analysis software. We then apply them to two historical marriage networks, one from the Republic of Venice and the other from Taiwan. By comparing centrality measures for the baseline and modified networks, we found that the most satisfactory of the proposed methods is the one that duplicates nodes of families with intra-family marriages and adds new edges that link these duplicated nodes to all the families to which the original node was connected. This procedure is computationally simple and conceptually sound, making it a useful tool for analysing marital networks.PID2020-115570GB-C22 (DemocratAI::UGR) Acceso abierto financiado por Universidad de Granada/CBU

Authors: He, M; Glasser, J; Pritchard, N; Bhamidi, S; Kaza, N

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Public Library of Science
• Journal: PLoS ONE
• Volume: 15
• Issue: 4
• Pages: e0230941-e0230941
• Publication date: 2020-04-29
• DOI: 10.1371/journal.pone.0230941
• EISSN: 1932-6203
• ISSN: 1932-6203
• Language: English
• Keywords: Urban agglomeration; Geography; Economic geography; Cartography; Regional science; Computer science; Network analysis; Metropolitan area