Journal of Civil Engineering
Vol 38, No 1 (2023)

LEARNING ABOUT CONCRETE-FILLED TUBE USING CHATGPT

Jen Hua Ling (University of Technology Sarawak, Malaysia)
Yong Tat Lim (University of Technology Sarawak, Malaysia)
Wen Kam Leong (University of Technology Sarawak, Malaysia)
How Teck Sia (University of Technology Sarawak, Malaysia)



Article Info

Publish Date
19 Jun 2023

Abstract

ChatGPT is an artificial intelligence that can understand context and respond appropriately. It could be used for research on a topic. However, ChatGPT does not always provide accurate information. Its performance has not been tested in engineering fields. In this study, ChatGPT was consulted about concrete-filled tubes (CFT), which is a structural element primarily subjected to axial load. Fifty-eight questions were posted to ChatGPT. ChatGPT’s responses (370 sentences) were evaluated. ChatGPT generated plagiarism-free statements, with only a 12% Turnitin similarity index. 78.6% of ChatGPT’s sentences were long and complex. Thus, Hemingway Editor gave them a Grade 14 for poor readability. The information given by ChatGPT can be classified as correct, erroneous, contradictory, and unverified. ChatGPT could be used as a research tool, but with limitations. It can explain the basic concepts of CFT but also provide inaccurate and contradictory information. A researcher needs to be cautious while using ChatGPT in research. ChatGPT could be used to test some hypotheses or theories. However, the quality of the output is dependent on the user’s critical inputs and an in-depth conversation with ChatGPT.

Copyrights © 2023






Journal Info

Abbrev

jce

Publisher

Subject

Description

Journal of Civil Engineering merupakan jurnal bidang teknik sipil yang mengacu pada sistem standar internasional dalam pengelolaannya dengan tujuan utama memajukan bidang teknik sipil melalui publikasi ilmiah demi terwujudnya kemudahan mendapatkan ilmu dan informasi serta mendukung kemajuan ...