Sensitivity to Evidence in Probabilistic Networks
Publication date
Authors
DOI
Document Type
Master Thesis
Metadata
Show full item recordCollections
License
CC-BY-NC-ND
Abstract
The probabilistic network framework is an approach to apply probability theory to reasoning with uncertainty in knowledge-based systems. Whereas researchers have studied the properties of parameter sensitivity analysis for probabilistic networks to quite some extent, evidence sensitivity analysis has received far less attention. The aim of my thesis is to present new, fundamental insights on sensitivity to evidence in probabilistic networks.
Keywords
bayesian networks, probabilistic networks, sensitivity analysis, probabilistic network pruning, evidence sensitivity set, evidence sensitivity function