Interval Based Evidence Sets

L.M. Rocha
Computer Research Group, MS P990
Los Alamos National Laboratory
Los Alamos, NM 87545

In: Proceedings of the ISUMA-NAFIPS'95. B. Ayyub (Ed.). IEEE Press. pp.624-629.

NOTE: The full version of this paper can be dowloaded in "zipped" postscript (.zip) format. Here only the abstract is shown due to the extensive use of mathematical symbols and figures. Notice that in some systems you may have to press the "shift" key while clicking on this link.

The relationship between set theory and the modeling of human categorization processes is the starting point of a discussion aiming at the definition of mathematical structures capable of better representing human categorization and linguistic capabilities. It is maintained that a theory of belief, which acknowledges subjectivity and observer dependence, is imperative in any attempt to model cognitive categorization. A connection between the Dempster-Shafer Theory of Evidence and Interval Valued Fuzzy Sets is proposed through the presentation of a structure called Evidence Set. This structure allows the introduction of all the recognized forms of uncertainty uncapturable by Fuzzy or Interval Valued Fuzzy Sets alone, as well as a representation of belief in a set arrangement.