Sensitivity to Evidence in Probabilistic Networks

Publication date

DOI

Document Type

Master Thesis

Collections

Open Access logo

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

Citation