Urban gender-based violence (GBV) is geographically concentrated, with recent research employing advanced geospatial machine learning to pinpoint high-risk areas accurately. A study published in Geomatics utilized open urban data from Valencia, Spain, revealing that factors like traffic intensity and socioeconomic conditions significantly influence GBV risk, more so than the presence of police stations. By categorizing data into a 25-meter grid and analyzing over 42,000 incidents, this research identifies critical predictors, such as high traffic volumes and nightlife density, while challenging assumptions about safety instigators like hospitals. The Random Forest model emerged as the most effective in predicting GBV hotspots, underscoring the impact of structural inequalities. These insights pave the way for targeted interventions, emphasizing traffic management and urban planning in violence prevention strategies. Highlighting the model’s interpretability, the study advocates for evidence-based urban policies that address the root causes of GBV, enhancing community safety effectively.
Share
Read more