Freedom: A Measure of Second-order Uncertainty for Intervalic Probability Schemes
This measure could be useful in knowledge engineering and expert systems, but it appears incremental as it builds on prior work.
The paper introduces a measure of second-order uncertainty called F, applicable to intervalic probability frameworks, possibility theory, and belief theory, arguing it indicates nonspecificity or freedom rather than ambiguity or vagueness.
This paper discusses a new measure that is adaptable to certain intervalic probability frameworks, possibility theory, and belief theory. As such, it has the potential for wide use in knowledge engineering, expert systems, and related problems in the human sciences. This measure (denoted here by F) has been introduced in Smithson (1988) and is more formally discussed in Smithson (1989a)o Here, I propose to outline the conceptual basis for F and compare its properties with other measures of second-order uncertainty. I will argue that F is an indicator of nonspecificity or alternatively, of freedom, as distinguished from either ambiguity or vagueness.