Sistema de inteligencia artificial basado en una red neuronal recurrente para la estimación de riesgo de violencia de género

The objective of this study is to establish the relevant information to identify, through an artificial intelligence (AI) system, which indicators increase the risk of experiencing gender-based violence. An AI system was implemented to assess the risk levels of physical, psychological, sexual, and economic violence to which a woman could be exposed, based on psychosocial factors derived from data collected from Mexican women in 2023. The methodology employed is heuristicand involves a recurrent neural network built in Python was trained with this data. The results obtained indicate that the cases of highest risk are women without a partner and those at the extremes of education levels, such as incomplete basic education or completed higher education. It can be concluded that it is possible to create a depersonalized database with quantifiable indicators to be used in future studies.