Assessing Socio-Economic Centrality Indicators in Urban Research:   A Bibliometric and systematic LLM-Assisted Review (2016–2026)

Khelil Cherfi Khadidja; Merzelkad Rym; Hourakhsh Ahmadnia; Hamza Benacer

doi:10.38027/ICCAUA2026EN0373

Authors

Khelil Cherfi Khadidja Department of Architecture, Institute of Architecture and urbansim, University of Blida 1, Blida, Algeria
Merzelkad Rym Department of Architecture, Institute of Architecture and urbansim, University of Blida 1, Blida, Algeria
Hourakhsh Ahmadnia Department of Architecture, Alanya Hamdullah Emin Pasa University, 07400 Alanya
Hamza Benacer Institute of Urban Techniques Management, Larbi Ben M’hidi University of Oum El Bouaghi, Oum El Bouaghi, Algeria

DOI:

https://doi.org/10.38027/ICCAUA2026EN0373

Keywords:

economic centrality, Bibliometric analysis, Large language models, Urban networks, Spatial analysis, Indicator assessment

Abstract

Socio-economic centrality indicators are widely used to explain the spatial concentration and interaction of economic activities, populations, and services within urban and regional systems. Between 2016 and 2026, rapid methodological diversification and interdisciplinary adoption have generated a fragmented and conceptually heterogeneous body of research. This paper presents a bibliometric and systematic review of socio-economic centrality indicators published between 2016 and 2026, combining classical bibliometric techniques with LLM assisted semantic synthesis. The combined approach enables the identification of dominant indicator families and their conceptual linkages. Eight socio-economic centrality indicators were identified: (1) accessibility and transport connectivity, (2) mobility and flow-based centrality, (3) economic and commercial activity, (4) amenities and urban attractors, (5) housing prices and investment dynamics, (6) population concentration and demographic indicators, (7) mixed-use development and activity intensity, and (8) innovation and knowledge networks. Results reveal a clear shift from traditional graph-based centrality measures toward multidimensional socio-spatial constructs that integrate these indicators, supported by GIS, big data, and machine learning. Persistent inconsistencies in indicator definition and operationalization are also observed. The study highlights the potential of LLM assisted reviews to enhance rigor, scalability, and interpretability in urban and socio-economic research.