Treatment of the information on gender-based violence. With contributions from artificial intelligence

Abstract

Introduction/objectivesViolence against women is still a serious social and health problem, despite the measures implemented in recent years. The examination of the victims by the forensic doctor in the courts is of great interest since it provides information related not only to the aggression, but also to their social, family, and economic environment. The objective is to use this information to identify groups at risk and improve/implement the necessary measures. Material and methodsIn this work, the forensic has collected, for 8 years, abundant data on the victims examined in l'Hospitalet de Llobregat. The sample includes 1622 cases of women who have been victims of gender violence. A descriptive study of the population and of the lesions has been carried out. ResultsThe paper presents the main variables studied, both socioeconomic and referring to the aggression itself. This study also analyzes the re-entry of the victims, the repetition of aggressions (revictimization), which are 10.9% of the sample. Finally, the results obtained after applying artificial intelligence techniques—in this case, CaRT classification trees—are presented. ConclusionsWith the results obtained, we conclude that the treatment of the information collected and systematized from the medical–forensic intervention allows a better understanding of Violence Against Women, from which we can extract suggestions on the adoption of care and support measures for the victims and the most vulnerable groups, as well as administrative resources and the optimization of prevention programs.