Case Law Analysis with Machine Learning in Brazilian Court

This paper aims the application of the knowledge discovering process in a judicial decisions database with the goal of unveiling the tendency of opinion Brazilian courts have, in relation to the favored party, employee or employer. Supervised machine learning techniques were used to classify documents. Such predictive model reached scores of more than 90% of ranking decision accuracy resulting in solid information about the tendency of each judge. Thus, this project contributes with a software that allows law firms to obtain information in a fast visual and exploratory way enabling them to focus and make more efforts in finding more effective legal strategies, than in jurisprudential research.