Coherent qualitative probability as an effective tool to represent uncertainty in the field of Artificial Intelligence is proposed. A suitable model to deal with vague and varying information is studied and some computable conditions are presented. An expert may introduce qualitative evaluations on a family of events containing only those strictly related to the problem. At any time the qualitative structure can be modified by referring to further events or relations or by better specifying the previously given ones. Coherence can be checked and (numerical) probabilistic statements, compatible with the qualitative judgements, can possibly be given.
Coherent qualitative probability and uncertainty in Artificial Intelligence / G., Coletti; Gilio, Angelo; Scozzafava, Romano. - STAMPA. - 1:(1990), pp. 132-138. (Intervento presentato al convegno 8th International Congress of Cybernetics and Systems tenutosi a New York City nel June 11-15, 1990).
Coherent qualitative probability and uncertainty in Artificial Intelligence
GILIO, ANGELO;SCOZZAFAVA, Romano
1990
Abstract
Coherent qualitative probability as an effective tool to represent uncertainty in the field of Artificial Intelligence is proposed. A suitable model to deal with vague and varying information is studied and some computable conditions are presented. An expert may introduce qualitative evaluations on a family of events containing only those strictly related to the problem. At any time the qualitative structure can be modified by referring to further events or relations or by better specifying the previously given ones. Coherence can be checked and (numerical) probabilistic statements, compatible with the qualitative judgements, can possibly be given.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
