| dc.rights.license | CC-BY-NC-ND | |
| dc.contributor.advisor | Renooij, Silja | |
| dc.contributor.author | Wojcik, Thomas | |
| dc.date.accessioned | 2023-05-25T02:01:02Z | |
| dc.date.available | 2023-05-25T02:01:02Z | |
| dc.date.issued | 2023 | |
| dc.identifier.uri | https://studenttheses.uu.nl/handle/20.500.12932/43929 | |
| dc.description.abstract | Naïve Bayesian Networks (NB) have been proven to be decently accurate classifiers, even in cases where their independency assumption does not hold. An approach to relax the independency assumption is to search through the possible single dependencies that can be added to the network, creating a so called Tree Augmented Bayesian Network (TAN), with the intention to improve the performance of the network. However, these TAN classifiers often perform about as good as a NB classifier, while increas | |
| dc.description.sponsorship | Utrecht University | |
| dc.language.iso | EN | |
| dc.subject | Naïve Bayesian Networks (NB) have been proven to be decently accurate classifiers, even in cases where their independency assumption does not hold. An approach to relax the independency assumption is to search through the possible single dependencies that can be added to the network, creating a so called Tree Augmented Bayesian Network (TAN), with the intention to improve the performance of the network. However, these TAN classifiers often perform about as good as a NB classifier, while increas | |
| dc.title | What explains the difference between naive Bayesian classifiers and tree-augmented Bayesian network classifiers. | |
| dc.type.content | Master Thesis | |
| dc.rights.accessrights | Open Access | |
| dc.subject.keywords | Bayesian Network, classification, Tree Augmented Bayesian Network, NB, TAN | |
| dc.subject.courseuu | Computing Science | |
| dc.thesis.id | 16915 | |
