Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
LEARNING ABOUT CONCRETE-FILLED TUBE USING CHATGPT
8
Zitationen
4
Autoren
2023
Jahr
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.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.719 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.628 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 8.176 Zit.
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
2019 · 6.880 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.