Evaluation of Uncertain Inference Models I: PROSPECTOR
Artificial Intelligence
2013-04-12 v1
Abstract
This paper examines the accuracy of the PROSPECTOR model for uncertain reasoning. PROSPECTOR's solutions for a large number of computer-generated inference networks were compared to those obtained from probability theory and minimum cross-entropy calculations. PROSPECTOR's answers were generally accurate for a restricted subset of problems that are consistent with its assumptions. However, even within this subset, we identified conditions under which PROSPECTOR's performance deteriorates.
Cite
@article{arxiv.1304.3117,
title = {Evaluation of Uncertain Inference Models I: PROSPECTOR},
author = {Robert M. Yadrick and Bruce M. Perrin and David S. Vaughan and Peter D. Holden and Karl G. Kempf},
journal= {arXiv preprint arXiv:1304.3117},
year = {2013}
}
Comments
Appears in Proceedings of the Second Conference on Uncertainty in Artificial Intelligence (UAI1986)