Scientific Collaboration in Research Networks: A Quantification Method by Using Gini Coefficient

In the scientific community, it is very common to try to create sound metrics for practically everything that can be measured. One of the current trends is to consider aspects from social networks for defining evaluation metrics. Following such a trend, our work proposes applying the Gini coefficient for evaluating research networks from two different perspectives. The first one analyzes the temporal evolution of research networks by considering the Gini coefficient of the distribution of researchers who have co-authored publications. The second one compares different internal collaboration networks of graduate programs and applies the Gini coefficient to support the ranking creation task. Both ideas are demonstrated through experiments that show the validity and applicability of our approach for quantifying scientific collaborations. Moreover, we also propose a new index that combines two metrics for evaluating the collaboration network of graduate programs. We believe that this index should be easily applied to other groups and research areas. This is the first time that the Gini coefficient is applied over social networks for evaluating research.