Joseph Makau, Mvurya Mgala, Obadiah Musau

ABSTRACT

Gender-Based Violence (GBV) is a serious issue that needs to be addressed by society and authorities with all available resources. Machine learning modeling has been used in many situations as an essential tool in addressing similar issues. Predictive models have been able to identify persons who are more likely to experience gender based violence by examining a variety of socio-demographic indicators, historical data, and contextual information.

This research uses the Snowballing methodology to review the literature. It involves identification of significant and recent literature on predictors of gender based violence using keywords such as gender, violence, prediction, data sets, and toxic relationships. Tracking of references and citations is conducted through backward and forward snowballing. To ensure efficiency and relevance of the reviewed papers to the topic of study, application of Boolean operators is utilized to reduce the search space. The review uses online databases including Google scholar.

The review results are the type of machine learning modeling techniques that have been used in in the prediction and prevention of gender based violence. The identified risk factors, patterns, and trends related to gender violence, and the analysis of a variety of data sources. Finally, the insights to the dynamics and underlying causes of violence and interventions and preventive measures.

The identified gap in this review motivates further research that will enhance the efforts to stop violence before it starts. Further research will be on the building of an optimized model that will act as a tool to reduce gender violence

Keywords: Machine Learning; modelling; Gender-Based violence.