Title :
Entropy of similarity relations in questionnaires and decision trees
Author :
Bouchon-Meunier, B. ; Yager, R.R.
Author_Institution :
LAFORIA, Paris VI Univ., France
Abstract :
An entropy measure is defined in the situation where the elements of a given universe have an underlying similarity relation in addition to a probability distribution. The authors study properties of this entropy corresponding to the use of several universes, and propose to apply this quantity to the construction of sequences of questions. They consider the general probabilistic case of questionnaires associated with probability distributions of answers to questions, and further suppose that similarity relations are given on the sets of answers or the universe
Keywords :
decision theory; entropy; probability; trees (mathematics); decision trees; entropy measure; general probabilistic case; probability distributions; questionnaires; similarity relations; Boolean algebra; Decision trees; Educational institutions; Entropy; Machine intelligence; Probability distribution; Strontium; Testing;
Conference_Titel :
Fuzzy Systems, 1993., Second IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0614-7
DOI :
10.1109/FUZZY.1993.327567
