Models of consensus for knowledge acquisition

Models of consensus are used in the knowledge acquisition process to determine what knowledge is built into a system. Unfortunately, in some cases, the consensus position is incorrect. This paper develops two analytic models of consensus that can be useful in the knowledge acquisition process. The first employs the binomial model to study the probability that using the consensus judgment for knowledge acquisition is correct or incorrect. That basic model is extended to account for both different levels of expertise and unequal prior odds. Conditions are found to indicate when the consensus model should be used in knowledge acquisition. The second model is a Bayesian model of the use of consensus judgment and knowledge acquisition. The Bayesian approach also finds that, in some cases, consensus judgment is not correct. Conditions similar to those for the binomial model are found to be appropriate for determining when the probability that the consensus is correct is greater than the probability that it is incorrect.