For some years now, there has been a growing concern for the adoption of gender perspectives in the judiciary. In this context, some studies have been carried out to analyze how stereotypes –which refer to the social and cultural construction of men and women based on their different physical, biological, sexual and social roles– affect judicial reasoning. In this paper, based on an interdisciplinary research project, it is proposed to use PLN techniques to detect these stereotypes in judicial decisions in a semi-automatic way. The objective is to assist in the review and analysis of rulings in order to detect situations and reasoning that evidence gender discrimination. Preliminary results obtained so far have identified sets of sentences typically associated with different stereotypes.