Causality Consistency Problem for Two Different Possibilistic Causal Models
Kôichi Yamada · 1999
Conditional Causal Probability (CCPR) and Conditional Causal Possibility (CCPO) have been proposed to express exact uncertainties of causalities, and some reasoning methods based on them have been studied to calculate probabilities or possibilities of unknown events under the condition that some events are known. CCPR/CCPO is a conditional probability/possibility of a causation event conditioned by its cause. Causation event is an Òevent that a cause actually causes an effect.Ó The paper proposes to classify causal models based on causation events into two types-- symmetrically valued and asymmetrically valued causal models-- depending on the properties of variables, which take an event as their value. It also shows the relations between CCPOs and conventional conditional possibilities in these two types of causal models. Then, it defines and discusses a Causality Consistency Problem, which is a problem to calculate the possibility of combined values of arbitrarily chosen unknown variables, when values of some other variables are known. 1.