Adewale Segun Alabi, Folasade Jokotade Odekunle, Oluwadamilola Ajoke Alabi, Babatunde Tolulope Adedeji, Kazeem Aderemi Bello, Rendani Wilson Maladzhi, Ilesanmi Daniyan and Adefemi Adeodu
Adv. Artif. Intell. Mach. Learn., 6 (3):5725-5745
1. Adewale Segun Alabi: Department of Architecture, Bells University of Technology Ota, Ogun State, Nigeria
2. Folasade Jokotade Odekunle: Department of Architecture, Bells University of Technology Ota, Ogun State, Nigeria
3. Oluwadamilola Ajoke Alabi: Department of Architecture, Bells University of Technology Ota, Ogun State, Nigeria
4. Babatunde Tolulope Adedeji: Institute of Systems Science, Durban University of Technology, Durban South Africa
5. Kazeem Aderemi Bello: Department of Mechanical Engineering, Durban University of Technology, Durban, South Africa
6. Rendani Wilson Maladzhi: Department of Mechanical Engineering, Durban University of Technology, Durban, South Africa
7. Ilesanmi Daniyan: Department of Mechatronics Engineering, Centre for Artificial Intelligence Bells University of Technology, P. M. B. 1015 Ota, Ogun State, Nigeria.
8. Adefemi Adeodu: Department of Project Management Centre for Artificial Intelligence Bells University of Technology, P. M. B. 1015 Ota, Ogun State, Nigeria.
DOI: 10.54364/AAIML.2026.63316
Article History: Received on: 14-Nov-25, Accepted on: 12-Jun-26, Published on: 19-Jun-26
Corresponding Author: Adewale Segun Alabi
Email: adewalesegunalabi12@gmail.com
Citation: Adewale Segun Alabi, et al. Evolution and Emerging Frontiers of Artificial Intelligence in Cultural Heritage Preservation: A 20-Year Bibliometric Review (2005-2024). Advances in Artificial Intelligence and Machine Learning.2026 (Ahead of Print) https://dx.doi.org/10.54364/AAIML.2026.63316
This study conducts a comprehensive bibliometric
review mapping the intellectual structure and evolution
of Artificial Intelligence (AI) in cultural heritage preservation (2005–2024).
Analysing 341 Scopus/Web of Science publications using VOSviewer, this study
employs performance analysis and science mapping (co-authorship, co-citation,
and keyword co-occurrence) in accordance with PRISMA guidelines. Three distinct
growth phases are identified: emergence (2005–2010), consolidation (2011–2017),
and acceleration (2018–2024). Foundational works bridged ecological modelling
and immersive technologies with modern AI. Thematic analysis reveals dominant
clusters: computer vision/3D reconstruction, deep learning/sensor fusion, and
hyperspectral material analysis. Emerging keywords ("physics-informed
machine learning," "transfer learning") signal shifts towards
domain-specific AI integration and cross-site generalization. The review
quantifies the field's evolution, collaboration networks, and conceptual shifts,
highlighting the interdisciplinary transfer of methodologies (e.g., from
ecology). It provides data-driven recommendations for future research,
including scalable AI frameworks and ethical guidelines. These contributions
advance Sustainable Development Goals (SDGs), particularly SDG 11.4
(safeguarding heritage) and SDG 4.7 (heritage education).
Â
Â