0.99 0.01 VisitToAsia 0.99 0.01 0.95 0.05 Smoke_OR_Abstain Smoker NonSmoker Smoking 0.99 0.01 0.9 0.1 Tuberculosis LungCancer Nothing CancerORTuberculosis CancerORTuberculosis CancerORTuberculosis TbOrCa 0.95 0.05 0.02 0.98 Smoking 0.7 0.3 0.4 0.6 TbOrCa Bronchitis 0.9 0.1 0.2 0.8 0.3 0.7 0.1 0.9 LungCancer Bronchitis Smoke_OR_Abstain 100 95 90 85 10 5 5 0 This is an example Bayesian network due to Lauritzen and Spiegelhalter, extended to an influence diagram. It is useful in demonstrating basics concepts of influence diagrams. The diagnostic part (the decision and the value node) helps in deciding whether or not to smoke.\nReference:\nThis is a modification (performed at BayesFusion) of an original network Asia, which first appeared in:\nLauritzen, Steffen L. & Spiegelhalter, David J. (1988). Local computations with probabilities on graphical structures and their application to expert systems, Journal of the Royal Statistical Society B, 50(2):157-224. Visit To Asia? 34 35 136 71 The node models whether the individual in question visited Asia recently. This is considered to be a risk factor in tuberculosis. Tuberculosis? 34 135 139 171 Presence or absence of tuberculosis. Smoke or Abstain 466 35 582 71 Should the individual smoke or abstain from smoking? Smoking? 264 35 353 71 Does the individual smoke or not? This is a serious risk factor in both lung cancer and in bronchitis. Lung Cancer? 183 135 281 171 Does the individual suffer from lung cancer? Tuberculosis or Lung Cancer? 122 226 255 286 Does the individual suffer from either tuberculosis or lung cancer? This node models practically existence of changes in the lungs, such as presence of condensed mass in the lungs. X-Ray Result 47 339 130 385 This node models the X-ray result. Both tuberculosis and lung cancer can be discovered on the X-ray because of presence of condensed mass in the lungs. Bronchitis? 402 237 488 273 Does the individual suffer from bronchitis? Dyspnea? 194 344 270 380 Does the individual suffer from dyspnea (shortness of breath)? Each of the diseases modeled can result in shortness of breath. Value 494 343 560 379 Total value to the individual. Both health and pleasure derived from smoking are factors in the total value. Please note that getting lung cancer is quite bad and this is reflected in the numbers. This is an example Bayesian network due to Lauritzen and Spiegelhalter, extended to an influence diagram. It is useful in demonstrating basics concepts of influence diagrams. The diagnostic part (the decision and the value node) helps in deciding whether or not to smoke.\n\nThe original network first appeared in:\nLauritzen, Steffen L. & Spiegelhalter, David J. (1988). Local computations with probabilities on graphical structures and their application to expert systems, Journal of the Royal Statistical Society B, 50(2):157-224. 619 20 928 324