作者: M Graf , K Nishijima , M H Faber
DOI: 10.1080/13287982.2009.11465008
关键词:
摘要: AbstractProbabilistic models are typically implemented into risk management systems using whatever relevant information is available prior to the implementation. However, in course of time more becomes and it significant practical importance be able update probabilistic based on new information. The present paper investigates a Bayesian approach for updating context natural hazards. networks proposed form basic tool representation knowledge uncertainties. Updating performed by instantiating variables corresponding observations from events This approach, however, necessitates that large can efficiently handled purpose compact object suggested. prop...