University of Twente Student Theses


Students' Trust in Automated Grading Through Explainable AI Visualizations.

Stoian, Vlad-Gabriel (2023) Students' Trust in Automated Grading Through Explainable AI Visualizations.

[img] PDF
Abstract:Currently, Artificial Intelligence (AI) has a big impact on people’s life. This technology evolved more and more every year, and now it is used worldwide, across most of the sectors. For example, its utilizations can be seen in transportation, logistics, medicine, military and also education. At present, different types of AI-based tools are developed for enhancing students’ experience in higher education. But, due to the increasing complexity of algorithms, lack of transparency can be observed, especially in AI-based tools used for summative assessments. Thus, it is believed that Explainable AI might play an important role in addressing this issue. The aim of this research is to discover what Explainable AI methods enhance the transparency of results in an AI-based grading tool and which of them are the most trusted by students. This will be done through literature review and interviews with university students. Furthermore, a clear overview of the selected Explainable AI methods will be provided with mock-ups. They will also be used to illustrate examples in the interviews. This research provides new insights for a major project at the University of Twente (an AI-grading tool).
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:50 technical science in general, 81 education, teaching
Programme:Business & IT BSc (56066)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page