scientific article; zbMATH DE number 3899959
Each modality of a fuzzy ordinal qualitative variable \(\beta\), is defined by a class \(\beta_{\lambda}\) of directed attributes, \(1\leq \lambda \leq \Lambda\). This definition makes use of a concrete notion of a degree of belonging of an individual to a profile of behaviour determined by a class of directed attributes. Accordingly to our general notion of proximity between variables, we construct here an association coefficient between an ordinary (said net by difference with the fuzzy case) qualitative ordinal variable and a fuzzy one. This coefficient generalizes those that we have established to compare two qualitative ordinal variables and which generalizes in a suitable way M. G. Kendall's coefficient. This new coefficient is able to evaluate the quality of a special type of an ordinal discrete regression between ordinal qualitative variables to be explained and an ordered sequence of classes issued from an automatic classification of a set of attributes-modalities.
- MDCgo takes up the association/correlation challenge for grouped ordinal data
- Fuzzy set covering of a set of ordinal attributes without parameter sharing
- Measures of association for fuzzy ordered categorical data
- scientific article; zbMATH DE number 3866402 (Why is no real title available?)
- scientific article; zbMATH DE number 3956883 (Why is no real title available?)
- scientific article; zbMATH DE number 4034885 (Why is no real title available?)
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3678482)