Local Pluralistic Homophily in Networks: A New Measure Based on Overlapping Communities

• Autores: Barraza, Fernando; Ramírez, Carlos; Fernández, Alejandro
• Año de publicación: 2023
• Idioma: inglés
• Tipo de recurso: documento de conferencia
• Materia: Ciencias de la Computación e Información; Local Pluralistic Homophily; Networks; Communities
• Institución: Comisión de Investigaciones Científicas de la Provincia de Buenos Aires

Pluralistic homophily is an important phenomenon in social network analysis as nodes tend to associate with others that share their same communities. In this work, we present the concept of local pluralistic homophily of a node in a network, along with a method to measure it. It is based on the assortativity index proposed by other authors. We analyze the distribution of local pluralistic homophily in different networks using publicly available datasets. We identify patterns of behavior of the proposed measure that relate to various structural and topological characteristics of a network. These findings are significant because they help better understand how pluralistic homophily affects communities. Furthermore, our results suggest possible applications of local pluralistic homophily in future research.