Large Language Models for Open Data Interoperability: Evidence, Gaps, and Practical Recommendations from a Systematic Literature Review
Digital Government: Research and Practice · 09.11.2026
DOI: 10.1145/3859875 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Digital Government: Research and Practice, online erschienen am 09.11.2026. Verfasst von Maria Ioanna Maratsi und 5 weiteren Autor:innen. Die Arbeit nennt 75 Literaturangaben.
Abstract (Originalsprache)
The accelerating growth of data generation and use in modern economies and societies has emphasised the need for interoperable data infrastructures and governance mechanisms that enable seamless data exchange, integration, and reuse across systems within the same or different sectors. Although open data has the potential to become a central component of public value creation, persistent challenges such as open data heterogeneous formats, inconsistent metadata, and limited semantic alignment, continue to constrain its utilisation. Recent advances in Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) offer new opportunities to address these issues by supporting automated classification and semantic enrichment of open data, as well as metadata generation, as well as cross-domain data harmonisation. Despite this potential of LLMs, research at the intersection of LLMs, interoperability, and open data remains fragmented and primarily concentrated in some isolated domains. To address this gap, a systematic literature review (SLR) was conducted to map and analyse existing research work employing LLMs to improve data interoperability within open data ecosystems. This literature review identifies the sectors in which LLM-based solutions are deployed for increasing data interoperability, the interoperability problems and challenges these solutions aim to address, as well as the models and methodological approaches adopted. Findings indicate that LLMs have strong potential to enhance semantic interoperability, improve metadata quality and data discovery, and support automated transformation of heterogeneous data sources. Moreover, the study includes emerging research directions regarding the future of LLMs as facilitators of open data interoperability and concludes by providing practical recommendations for the development of AI-enabled interoperable open data infrastructures.
Zitieren
APA
Maratsi, M. I., Ali, M., Papageorgiou, G., Alexopoulos, C., Loukis, E., & Charalabidis, Y. (2026). Large Language Models for Open Data Interoperability: Evidence, Gaps, and Practical Recommendations from a Systematic Literature Review. Digital Government: Research and Practice. https://doi.org/10.1145/3859875
Vancouver
Maratsi MI, Ali M, Papageorgiou G, Alexopoulos C, Loukis E, Charalabidis Y. Large Language Models for Open Data Interoperability: Evidence, Gaps, and Practical Recommendations from a Systematic Literature Review. Digital Government: Research and Practice. 2026. doi:10.1145/3859875
BibTeX
@article{Maratsi2026,
author = {Maratsi, Maria Ioanna and Ali, Mohsan and Papageorgiou, Georgios and Alexopoulos, Charalampos and Loukis, Euripidis and Charalabidis, Yannis},
title = {Large Language Models for Open Data Interoperability: Evidence, Gaps, and Practical Recommendations from a Systematic Literature Review},
journal = {Digital Government: Research and Practice},
year = {2026},
doi = {10.1145/3859875}
}
EUVIMED