Accession Number:

ADA224305

Title:

Combining Stochastic Uncertainty and Linguistic Inexactness: Theory and Experimental Evaluation of Four Fuzzy Probability Models

Descriptive Note:

Rept. for Jun 1982-Sep 1986,

Corporate Author:

NORTH CAROLINA UNIV AT CHAPEL HILL DEPT OF PSYCHOLOGY

Personal Author(s):

Report Date:

1989-01-01

Pagination or Media Count:

45.0

Abstract:

Two major sources of imprecision in human knowledge are identified in this reprint linguistic inexactness and stochastic uncertainty. The study argues that since in most realistic situations, both exist simultaneously, they should both be combined in a frugal framework in order to yield realistic solutions, and presents such a framework by combining concepts from probability and fuzzy set theories. Four models that attempt to account for linguistic or numeric responses in various probability elicitation tasks were tested in such a framework. The linguistic models were relatively effective in predicting subject responses compared to a random choice model, but the numeric model was not. These and other results suggest that subjects are unable to represent the full complexity of a problem, instead they adopt a simplified view of the problem, representing vague linguistic concepts by multiple-crisp representations the level sets. All mental computation is done at these surrogate levels. kr

Subject Categories:

  • Psychology
  • Statistics and Probability

Distribution Statement:

APPROVED FOR PUBLIC RELEASE