Maintenance in Probabilistic Knowledge-Based Systems

TL;DR

This paper explores methods for maintaining probabilistic knowledge in Bayesian networks, focusing on reducing the expert's effort during updates by leveraging conditional independencies and special cases.

Contribution

It introduces specific cases where existing probabilities do not need reassessment, improving the efficiency of knowledge base maintenance in probabilistic systems.

Findings

Identification of three special cases reducing probability reassessment

Use of conditional independencies to limit updates

Enhanced efficiency in probabilistic knowledge maintenance

Abstract

Recent developments using directed acyclical graphs (i.e., influence diagrams and Bayesian networks) for knowledge representation have lessened the problems of using probability in knowledge-based systems (KBS). Most current research involves the efficient propagation of new evidence, but little has been done concerning the maintenance of domain-specific knowledge, which includes the probabilistic information about the problem domain. By making use of conditional independencies represented in she graphs, however, probability assessments are required only for certain variables when the knowledge base is updated. The purpose of this study was to investigate, for those variables which require probability assessments, ways to reduce the amount of new knowledge required from the expert when updating probabilistic information in a probabilistic knowledge-based system. Three special cases (ignored outcome, split outcome, and assumed constraint outcome) were identified under which many of the original probabilities (those already in the knowledge-base) do not need to be reassessed when maintenance is required.