Package: triversity 2.0

triversity: Diversity Measures on Multipartite Graphs

Computing diversity measures on multipartite graphs. This package first implements a parametrized family of such diversity measures which apply on probability distributions. Sometimes called "True Diversity", this family contains famous measures such as the richness, the Shannon entropy, the Herfindahl-Hirschman index, and the Berger-Parker index. Second, the package allows to apply these measures on probability distributions resulting from random walks between the parts of multipartite graphs. By defining an initial distribution on a given part of the graph and a path to follow between the different parts, the probability of the walker's position within the final part is then computed, thus providing a particular instance of diversity to measure.

Authors:Robin Lamarche-Perrin [aut, cre]