Subjective Logic has operators for conditional deduction a nd conditional abduction where subjective opinions are input arguments. With these operators traditional Baye sian reasoning can be generalised from taking only probabilistic arguments to also taking opinions as arguments, th ereby allowing Bayesian modeling of situations affected by uncertainty and incomplete information. Conditional dedu ction is a Bayesian reasoning process that goes in the same direction as that of the input argument conditionals, where as conditional abduction is a Bayesian reasoning process th at goes in the direction opposite to that of the input argument c onditionals. Conditional abduction is in fact a two-step process that first involves the computation of inverted cond itionals, and then conditional deduction based on the inver ted conditionals. This paper describes an improved generalize d m thod for inverting opinion conditionals in order to supp ort general Bayesian reasoning in subjective logic.
[1]
J. Pearl.
Causality: Models, Reasoning and Inference
,
2000
.
[2]
Audun Jøsang,et al.
Interpretation and fusion of hyper opinions in subjective logic
,
2012,
2012 15th International Conference on Information Fusion.
[3]
Audun Jøsang,et al.
Conditional Reasoning with Subjective Logic
,
2009,
J. Multiple Valued Log. Soft Comput..
[4]
J. Koehler.
The base rate fallacy reconsidered: Descriptive, normative, and methodological challenges
,
1996,
Behavioral and Brain Sciences.
[5]
Audun Jøsang,et al.
Conditional Deduction Under Uncertainty
,
2005,
ECSQARU.
[6]
Audun Jøsang.
Abductive Reasoning with Uncertainty
,
2008
.