top of page

[實習主題] -- Revolutionizing Safety Analytics - Harnessing the Power of Large Language Models for Safety Report Information Extraction

指導老師: 林之謙 教授

專案經理: 黎芳玲

實習名額: 1位

是否同意遠距: 是

專案類型: 應用型 (預期讓實習生把技能運用在正在進行的研究計劃中)


Safety management is a classic problem in construction project management. The issue of ensuring safety on construction sites by applying digital technologies has received considerable critical attention. Digital technologies have brought about a transformative wave in addressing these concerns, focusing on leveraging Natural Language Processing (NLP) methods. The past decade has witnessed the rapid development of the Large Language Model (LLM) in various applications, such as question and answering systems, text classification, etc.


The present project stands at the intersection of construction site safety and cutting-edge technology. It recognizes the critical role of site engineers, who gather and document crucial information recorded within safety reports daily. These reports, often comprising detailed accounts of observations, hazards, safety measures, and incident reports, serve as invaluable data repositories.


This project utilizes NLP by creating a Large Language Model to extract and analyze information contained in safety reports. The primary objectives include the classification of reports and the systematic storage of safety data in a database. This strategic approach facilitates the efficient organization of information and sets the stage for further extracting pertinent insights during the safety management process.



30 次查看0 則留言
bottom of page