University of Twente Student Theses

Login

Leveraging Generative AI for Educational Feedback in Data Modeling within Information Systems

Petrou, M. (2024) Leveraging Generative AI for Educational Feedback in Data Modeling within Information Systems.

[img] PDF
393kB
Abstract:The demand for IT talent has increased compoundly in recent years across many industries, leading to the rise of citizen developers. Citizen or novice developers are individuals without formal expertise in IT, aiming to fill this gap. This research explores the capabilities of generative AI technologies, specifically ChatGPT-4, to evaluate the work of novices and provide feedback in the context of data and behavior modeling. The study assessed ChatGPT-4’s feedback based on four predefined criteria which are the following: accuracy, relevance, comprehensiveness, and adherence to standards. After conducting a survey and receiving the opinion of many experienced participants in data modeling, results showed that ChatGPT-4 can provide valuable and relevant feedback, helping inexperienced developers such as the novices enhance their models. However, in some cases it struggled to identify the correct relationships of classes within data models and sometimes offered feedback that was too generic. Future studies relevant to this topic should involve more experts from across different contexts and use models that are more complex to challenge further the limitations of this tool. The enormous potential of this tool requires continuous improvement and research around it, to maximize effectiveness in helping novices learn.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:https://purl.utwente.nl/essays/101019
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page