Combining Multiple-Valued Logics in Modular Expert Systems
This addresses a modularity issue in expert systems for AI researchers, but it appears incremental as it builds on existing multiple-valued logic frameworks.
The paper tackles the problem of enabling communication between modules in a large knowledge-based system when each module uses a different uncertainty calculus, by analyzing it as an inference-preserving communication problem restricted to truth-functional multiple-valued logics.
The way experts manage uncertainty usually changes depending on the task they are performing. This fact has lead us to consider the problem of communicating modules (task implementations) in a large and structured knowledge based system when modules have different uncertainty calculi. In this paper, the analysis of the communication problem is made assuming that (i) each uncertainty calculus is an inference mechanism defining an entailment relation, and therefore the communication is considered to be inference-preserving, and (ii) we restrict ourselves to the case which the different uncertainty calculi are given by a class of truth functional Multiple-valued Logics.