In the past 30 years, technological advances in computation have allowed to storage, process, and analyze massive amounts of data resulting from all sorts of human activities. The ability to address these large-scale data is crucial in developing new understandings of the sociological and cultural aspects underlying these human activities. In the case of legal studies, digital resources from court and legislative activities (such as legal codes and judicial decisions) can be easily accessed in public repositories. Although the legal domain does not rely in computational and quantitative approaches as much as other fields do, the use of such techniques has increased significantly over the years, with many scholars exposing the benefits of adopting empirical and quantitative methodologies to generate objective, falsifiable and reproducible knowledge. In the present thesis, we use network science and statistical inference tools over largescale corpora of judicial decisions to reveal and understand patterns behind the functioning of the judicial system