Datasets for DEXA 2015 Paper

Analyzing the Strength of Co-authorship Ties with Neighborhood Overlap

Michele A. Brandão, Mirella M. Moro
Departamento de Ciência da Computação
Universidade Federal de Minas Gerais, Belo Horizonte, MG
{micheleabrandao,mirella}@dcc.ufmg.br


Abstract

Evaluating researchers' scientific productivity usually relies on bibliometry only, which may not be always fair. Here, we take a step forward on analyzing such data by exploring the strength of coauthorship ties in social networks. Specifically, we build co-authorship social networks by extracting the datasets of three research areas (sociology, medicine and computer science) from a real digital library and analyze how topological properties relate to the strength of ties. Our results show that different topological properties explain variations in the strength of co-authorship ties, depending on the research area. Also, we show that neighborhood overlap can be applied to scientific productivity evaluation and analysis beyond bibliometry.

Datasets

Links to the complete datasets used in the experiments: