Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | English |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.19707320 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>Bangladesh's land administration system, encumbered by colonial-era manual records, fragmented institutional<br>mandates, and a backlog of more than 3.6 million pending land disputes, stands at an inflection point. This<br>review synthesises peer-reviewed literature, policy documents, and pilot implementations published between<br>2015 and 2024 to evaluate the practical application of Artificial Intelligence (AI) to land management in<br>Bangladesh. Six AI capabilities are examined: (i) optical character recognition (OCR) and natural-language<br>processing of legacy mutation records, (ii) satellite-image segmentation for parcel boundary extraction, (iii)<br>machine-learning fraud detection in registration workflows, (iv) blockchain-anchored title management, (v)<br>chatbot-mediated citizen services, and (vi) predictive analytics for char land and climate displacement.<br>Drawing on the a2i Smart Bangladesh 2041 roadmap and comparative cases from India, Rwanda, and Estonia,<br>the paper proposes a five-layer implementation framework — data, model, governance, service, and trust —<br>and advances eight policy recommendations covering data standards, sandboxed pilots, workforce reskilling,<br>and statutory protection against algorithmic bias. The review concludes that AI is not a substitute for<br>institutional reform, but, when embedded within a rights-based digital public infrastructure, can compress<br>mutation timelines from months to days, reduce fraud exposure, and restore citizen trust in one of South Asia's<br>most contested public services.<br>Keywords: Artificial Intelligence; Land Management; Bangladesh; Digital Land Records; Geospatial AI;<br>e-Governance; Land Registry; Smart Bangladesh 2041</p>