Expert System Consultation Control Strategy

Abstract

User interfaces to expert systems represent a bottleneck since consultation time is proportional to the amount of information the system asks the user to supply. An efficient, rather than exhaustive, strategy to direct user questioning will reduce consultation time and effort. An intelligent strategy to minimize questioning, the merit system, has been successfully implemented in Battle, an expert consultant system developed for the Marine Corps. The merit strategy enables Battle to focus the consultation process on the most meritorious questions allowing the military commander to respond quickly awith the most pertinent information. The merit system, originally denned for logical functions in the Multiple program, has been extended to the Mycin style of propagation and to the method of subjective Bayesian assignments used by Prospector. A procedure for merit calculations with any differentiable, real-valued assignment function is presented. Our experience has shown that merit values provide an efficient flow of control for expert consultation.

Cite

Text

Slagle and Gaynor. "Expert System Consultation Control Strategy." AAAI Conference on Artificial Intelligence, 1983.

Markdown

[Slagle and Gaynor. "Expert System Consultation Control Strategy." AAAI Conference on Artificial Intelligence, 1983.](https://mlanthology.org/aaai/1983/slagle1983aaai-expert/)

BibTeX

@inproceedings{slagle1983aaai-expert,
  title     = {{Expert System Consultation Control Strategy}},
  author    = {Slagle, James R. and Gaynor, Michael W.},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {1983},
  pages     = {369-372},
  url       = {https://mlanthology.org/aaai/1983/slagle1983aaai-expert/}
}