Extracting Structured Requirements from Unstructured Building Technical Specifications for Building Information Modeling

TL;DR

This paper presents a method combining NLP and BIM to automatically extract structured requirements from unstructured French technical specifications, using transformer models and transfer learning.

Contribution

It introduces a novel approach integrating transformer-based NLP models with BIM for extracting requirements from French construction documents, with benchmark comparisons of multiple models.

Findings

CamemBERT and Fr_core_news_lg achieved over 90% F1 in NER.

Random Forest achieved over 80% F1 in Relation Extraction.

Models outperform rule-based and traditional methods.

Abstract

This study explores the integration of Building Information Modeling (BIM) with Natural Language Processing (NLP) to automate the extraction of requirements from unstructured French Building Technical Specification (BTS) documents within the construction industry. Employing Named Entity Recognition (NER) and Relation Extraction (RE) techniques, the study leverages the transformer-based model CamemBERT and applies transfer learning with the French language model Fr\_core\_news\_lg, both pre-trained on a large French corpus in the general domain. To benchmark these models, additional approaches ranging from rule-based to deep learning-based methods are developed. For RE, four different supervised models, including Random Forest, are implemented using a custom feature vector. A hand-crafted annotated dataset is used to compare the effectiveness of NER approaches and RE models. Results indicate that CamemBERT and Fr\_core\_news\_lg exhibited superior performance in NER, achieving F1-scores over 90\%, while Random Forest proved most effective in RE, with an F1 score above 80\%. The outcomes are intended to be represented as a knowledge graph in future work to further enhance automatic verification systems.