This project looks at predicting Gender-based Violence (GBV) with the ACLED (Armed Conflict Location and Event Data) dataset. GBV remains a critically under-addressed crisis, particularly in conflict-affected states where justice systems are weak; and support structures for victims are limited. By applying machine learning classifications, including Support Vector Machines (SVM) and Random Forest (RF), this project aims to identify patterns in violent events using descriptive notes from conflict datasets targeting women, to contribute to improved monitoring and response efforts for political GBV.
political GBV.ipynb - this script includes data wrangling and models performing.
Data from Armed Conflict Location & Event Data Project (ACLED); www.acleddata.com
April 2024 to March 2025 ACLED data are filtered for this project.
Raleigh, C., Kishi, R. & Linke, A. Political instability patterns are obscured by conflict dataset scope conditions, sources, and coding choices. Humanit Soc Sci Commun 10, 74 (2023). https://doi.org/10.1057/s41599-023-01559-4
Sunny Ren