Alves, Daniella (2026) Uncover to Discover: A Review of Explainable Artificial Intelligence for Knowledge Discovery. Bachelor's Thesis, Artificial Intelligence.
|
Text
BScThesisDaniellaAlves.pdf Download (555kB) | Preview |
Abstract
Explainable Artificial Intelligence (XAI) advocates for insight into models' reasoning through interpretable models and post-hoc explanation methods. With existing understanding of XAI used to justify, control, and improve, there is room to elaborate on how XAI has been utilised for the purpose of discovery. This study synthesises 42 research papers focusing on XAI for knowledge discovery, exploring the range of methods used. Furthermore, a taxonomy of discovery level is developed to analyse the findings and novelty potential of the examined studies. In doing so, this review assesses the landscape of XAI development, noting the ways in which XAI facilitates discovery, the potential risks of unguarded trust in XAI methods, and proposing to view discovery as a spectrum rather than a restricted definition.
| Item Type: | Thesis (Bachelor's Thesis) |
|---|---|
| Supervisor name: | Schoeffer, J.J. |
| Degree programme: | Artificial Intelligence |
| Thesis type: | Bachelor's Thesis |
| Language: | English |
| Date Deposited: | 28 Jul 2026 06:37 |
| Last Modified: | 28 Jul 2026 06:37 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/37464 |
Actions (login required)
 |
View Item |