Mapping the AECO Research Landscape using Topic Modeling, Bibliometrics and Information Extraction Methods

Journal article

Vanni Zavarella, Juan Gamero-Salinas, Danilo Dessí, Sergio Consoli, Gianni Fenu, Diego Reforgiato Recupero
IEEE Access, 2026, pp. 1-1

DOI: 10.1109/ACCESS.2026.3688923

Cite

APA Click to copy
Zavarella, V., Gamero-Salinas, J., Dessí, D., Consoli, S., Fenu, G., & Recupero, D. R. (2026). Mapping the AECO Research Landscape using Topic Modeling, Bibliometrics and Information Extraction Methods. IEEE Access, 1–1. https://doi.org/10.1109/ACCESS.2026.3688923

Chicago/Turabian Click to copy
Zavarella, Vanni, Juan Gamero-Salinas, Danilo Dessí, Sergio Consoli, Gianni Fenu, and Diego Reforgiato Recupero. “Mapping the AECO Research Landscape Using Topic Modeling, Bibliometrics and Information Extraction Methods.” IEEE Access (2026): 1–1.

MLA Click to copy
Zavarella, Vanni, et al. “Mapping the AECO Research Landscape Using Topic Modeling, Bibliometrics and Information Extraction Methods.” IEEE Access, 2026, pp. 1–1, doi:10.1109/ACCESS.2026.3688923.

BibTeX Click to copy

@article{zavarella2026a,
  title = {Mapping the AECO Research Landscape using Topic Modeling, Bibliometrics and Information Extraction Methods},
  year = {2026},
  journal = {IEEE Access},
  pages = {1-1},
  doi = {10.1109/ACCESS.2026.3688923},
  author = {Zavarella, Vanni and Gamero-Salinas, Juan and Dessí, Danilo and Consoli, Sergio and Fenu, Gianni and Recupero, Diego Reforgiato}
}

Abstract
In this work, we address the challenge of mapping the major areas and trending topics in the Architecture, Engineering, Construction, and Operations (AECO) research domain by integrating topic modeling with bibliometric analysis and information extraction techniques. After collecting a large corpus of AECO research papers through the OpenAlex public API, we optimize a neural topic modeling algorithm to identify research macro topics, leveraging quality control from domain experts. We then apply bibliometric tools to detect key institutions in each research area and customize an LLM-powered information extraction pipeline to identify emerging research problems and solution trends. All extracted entities and relationships are represented in a knowledge graph, enabling structured queries and semantic exploration. Finally, we describe an interactive visualization dashboard designed to support data analytics across the AECO research landscape.