Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance Regulations

Automated compliance checking (ACC) in the Architecture, Engineering, and Construction (AEC) sector represents a pivotal task which is traditionally executed manually, demanding significant time and labor. This work investigates the automation of the Requirement, Applicability, Selection, and Except...

Full description

Saved in:
Bibliographic Details
Main Authors: Dhoyazan Al-Turki, Hansi Hettiarachchi, Mohamed Medhat Gaber, Mohammed M. Abdelsamea, Shadi Basurra, Sima Iranmanesh, Hadeel Saadany, Edlira Vakaj
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10778545/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850106635436949504
author Dhoyazan Al-Turki
Hansi Hettiarachchi
Mohamed Medhat Gaber
Mohammed M. Abdelsamea
Shadi Basurra
Sima Iranmanesh
Hadeel Saadany
Edlira Vakaj
author_facet Dhoyazan Al-Turki
Hansi Hettiarachchi
Mohamed Medhat Gaber
Mohammed M. Abdelsamea
Shadi Basurra
Sima Iranmanesh
Hadeel Saadany
Edlira Vakaj
author_sort Dhoyazan Al-Turki
collection DOAJ
description Automated compliance checking (ACC) in the Architecture, Engineering, and Construction (AEC) sector represents a pivotal task which is traditionally executed manually, demanding significant time and labor. This work investigates the automation of the Requirement, Applicability, Selection, and Exception (RASE) methodology for building regulatory compliance through the utilization of Large Language Models (LLMs) and active learning techniques. Specifically, we focus on the development and assessment of a system using the OpenAI GPT-4o model to transmute building regulation texts into structured YAML formats conducive to ACC processes. The study encompasses three experimental paradigms: few-shot learning, fine-tuning learning, and progressive active learning. Initial results from the few-shot learning experiment illustrate the model’s preliminary ability to interpret and process regulatory texts with limited examples. Fine-tuning enhances model performance by training it on a specialized dataset, thereby improving structural and textual accuracy. Progressive active learning, by iteratively incorporating expert feedback, further refines the accuracy of the model. The findings demonstrate substantial enhancements in both structural and semantic accuracies of the generated YAML files, underscoring the potential of integrating LLMs with active learning to streamline regulatory compliance automation. The methodologies and results presented here offer a comprehensive framework for advancing future research and practical applications in the domain of automated regulatory compliance.
format Article
id doaj-art-6d434dc84fec47069cc040a7f9068588
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-6d434dc84fec47069cc040a7f90685882025-08-20T02:38:46ZengIEEEIEEE Access2169-35362024-01-011218529118530610.1109/ACCESS.2024.351243410778545Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance RegulationsDhoyazan Al-Turki0Hansi Hettiarachchi1https://orcid.org/0000-0003-4609-5001Mohamed Medhat Gaber2https://orcid.org/0000-0003-0339-4474Mohammed M. Abdelsamea3https://orcid.org/0009-0000-5677-6818Shadi Basurra4Sima Iranmanesh5Hadeel Saadany6https://orcid.org/0000-0002-2620-1842Edlira Vakaj7School of Computing and Digital Technology, Birmingham City University, Birmingham, U.K.School of Computing and Communications, Lancaster University, Lancaster, U.K.School of Computing and Digital Technology, Birmingham City University, Birmingham, U.K.Department of Computer Science, University of Exeter, Exeter, U.K.School of Computing and Digital Technology, Birmingham City University, Birmingham, U.K.School of Computing and Digital Technology, Birmingham City University, Birmingham, U.K.School of Computing and Digital Technology, Birmingham City University, Birmingham, U.K.School of Computing and Digital Technology, Birmingham City University, Birmingham, U.K.Automated compliance checking (ACC) in the Architecture, Engineering, and Construction (AEC) sector represents a pivotal task which is traditionally executed manually, demanding significant time and labor. This work investigates the automation of the Requirement, Applicability, Selection, and Exception (RASE) methodology for building regulatory compliance through the utilization of Large Language Models (LLMs) and active learning techniques. Specifically, we focus on the development and assessment of a system using the OpenAI GPT-4o model to transmute building regulation texts into structured YAML formats conducive to ACC processes. The study encompasses three experimental paradigms: few-shot learning, fine-tuning learning, and progressive active learning. Initial results from the few-shot learning experiment illustrate the model’s preliminary ability to interpret and process regulatory texts with limited examples. Fine-tuning enhances model performance by training it on a specialized dataset, thereby improving structural and textual accuracy. Progressive active learning, by iteratively incorporating expert feedback, further refines the accuracy of the model. The findings demonstrate substantial enhancements in both structural and semantic accuracies of the generated YAML files, underscoring the potential of integrating LLMs with active learning to streamline regulatory compliance automation. The methodologies and results presented here offer a comprehensive framework for advancing future research and practical applications in the domain of automated regulatory compliance.https://ieeexplore.ieee.org/document/10778545/Automated compliance checkingRASElarge language modelsactive learning
spellingShingle Dhoyazan Al-Turki
Hansi Hettiarachchi
Mohamed Medhat Gaber
Mohammed M. Abdelsamea
Shadi Basurra
Sima Iranmanesh
Hadeel Saadany
Edlira Vakaj
Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance Regulations
IEEE Access
Automated compliance checking
RASE
large language models
active learning
title Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance Regulations
title_full Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance Regulations
title_fullStr Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance Regulations
title_full_unstemmed Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance Regulations
title_short Human-in-the-Loop Learning With LLMs for Efficient RASE Tagging in Building Compliance Regulations
title_sort human in the loop learning with llms for efficient rase tagging in building compliance regulations
topic Automated compliance checking
RASE
large language models
active learning
url https://ieeexplore.ieee.org/document/10778545/
work_keys_str_mv AT dhoyazanalturki humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations
AT hansihettiarachchi humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations
AT mohamedmedhatgaber humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations
AT mohammedmabdelsamea humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations
AT shadibasurra humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations
AT simairanmanesh humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations
AT hadeelsaadany humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations
AT edliravakaj humaninthelooplearningwithllmsforefficientrasetagginginbuildingcomplianceregulations