Practical Issues in Constructing a Bayes' Belief Network
This work tackles the problem of efficiently constructing probabilistic models for experts in fields like AI and decision-making, but it is incremental as it builds on existing network concepts.
The paper addresses the practical challenges in building Bayes' belief networks to represent expert knowledge, proposing techniques like generalizations of the 'noisy OR gate' and sensitivity analysis to guide structuring and quantifying uncertain relationships, with results indicating when rough probability assessments suffice versus when precision is needed.
Bayes belief networks and influence diagrams are tools for constructing coherent probabilistic representations of uncertain knowledge. The process of constructing such a network to represent an expert's knowledge is used to illustrate a variety of techniques which can facilitate the process of structuring and quantifying uncertain relationships. These include some generalizations of the "noisy OR gate" concept. Sensitivity analysis of generic elements of Bayes' networks provides insight into when rough probability assessments are sufficient and when greater precision may be important.