A proposal for the EI index for fuzzy groups

In this article, a measure that quantifies the relational structure within and between groups is proposed, comprising not only analysis of disjoint or non-disjoint groups, but also fuzzy groups. This based on existing known as EI index. The current index homophily applied to networks with presence although large scale rarely exist in many empirical networks. addition, combination edge node weights evaluation proposed. We tested two different contexts. first co-authorship network, where researchers, actors are divided according time Ph.D. completion. second network formed by trade relations countries American continent, grouped Human Development Index. application proposed these justified imprecision information difficulty allocating nodes specific group, being necessary define affiliation levels. Therefore, new allows expanding social networks, for types attributes, thus generating previously unexplored knowledge.