Learn R Programming

bnRep (version 0.0.3)

humanitarian: humanitarian Bayesian Network

Description

You only derive once (YODO): Automatic differentiation for efficient sensitivity analysis in Bayesian networks.

Arguments

Value

An object of class bn.fit. Refer to the documentation of bnlearn for details.

Format

A discrete Bayesian network to assess the country-level risk associated with humanitarian crises and disasters. The Bayesian network is learned as in the referenced paper. The vertices are:

RISK

(low, medium, high);

EARTHQUAKE

(low, medium, high);

FLOOD

(low, medium, high);

TSUNAMI

(low, medium, high);

TROPICAL_CYCLONE

(low, medium, high);

DROUGHT

(low, medium, high);

EPIDEMIC

(low, medium, high);

PCR

Projected conflict risk (low, medium, high);

CHVCI

Current highly violent conflict intensity (low, medium, high);

D_AND_D

Development and deprivation (low, medium, high);

ECON_DEP

Economic dependency (low, medium, high);

UNP_PEOPLE

Unprotected people (low, medium, high);

HEALTH_COND

Health conditions (low, medium, high);

CHILDREN_U5

(low, medium, high);

RECENT_SHOCKS

(low, medium, high);

FOOD_SECURITY

(low, medium, high);

OTHER_VULN_GROUPS

Other vulnerable groups (low, medium, high);

GOVERNANCE

(low, medium, high);

COMMUNICATION

(low, medium, high);

PHYS_INFRA

Physical infrastructure (low, medium, high);

ACCESS_TO_HEALTH

(low, medium, high);

References

Ballester-Ripoll, R., & Leonelli, M. (2022, September). You only derive once (YODO): automatic differentiation for efficient sensitivity analysis in Bayesian networks. In International Conference on Probabilistic Graphical Models (pp. 169-180). PMLR.