Generating process anomalies using a taxonomy of fraud characteristics and Markov models for accurate detection
| dc.rights.license | CC-BY-NC-ND | |
| dc.contributor.advisor | Lu, X. | |
| dc.contributor.author | Veldman, Jochem | |
| dc.date.accessioned | 2022-09-09T03:01:52Z | |
| dc.date.available | 2022-09-09T03:01:52Z | |
| dc.date.issued | 2022 | |
| dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/42635 | |
| dc.description.sponsorship | Utrecht University | |
| dc.language.iso | EN | |
| dc.subject | Supervised models perform much better than unsupervised models in detecting anomalies accurately. However, supervised models can often not be chosen since there is a lack of labelled data. Labelling data is a time-consuming and expensive task, so alternatives are wanted. Researchers have developed ways to work around the lack of labels, such as under- and oversampling. However, these methods are based on assumptions, the models trained with the generated data do not always show promising results | |
| dc.title | Generating process anomalies using a taxonomy of fraud characteristics and Markov models for accurate detection | |
| dc.type.content | Master Thesis | |
| dc.rights.accessrights | Open Access | |
| dc.subject.courseuu | Business Informatics | |
| dc.thesis.id | 9647 |
