Logic, Reasoning and A Programming Language for Simulating Economic and Business Processes with Artificially Intelligent Agents

2 Modelling Qualitative Information

It has been recognized for some time that a difference between formal logics and Bayesian methods is that the former deals more naturally with qualitative information and with limitations on the abilities of decision-makers to identify all possible outcomes or even completely to specify their own preference functions*1.

This is because formal logical languages can also include a universe of user-defined objects, functions and predicates which can directly represent such information in a natural way. Thus the formal logic can be easily extended with the addition of such terms in a very flexible and expressive way. The formal structure of the logic is thus the "glue" that is used to relate these, in a similar manner that Bayesian theory relates probabilities. Perhaps the ultimate indication of the expressivity of a logical formalism is that it can be used to formalise Bayesian theory fairly easily, where this is not possible the other way around.

Logic, Reasoning and A Programming Language for Simulating Economic and Business Processes with Artificially Intelligent Agents - 12 APR 96

[Next] [Previous] [Top] [Contents]

Generated with CERN WebMaker