Predicting Lethal Outcomes of Protests
Protests have been a powerful force for social change throughout history, but unfortunately, they can sometimes lead to tragic outcomes. This project aims to predict how governments would respond to mass mobilization protests, of which that results in shootings or killings, with the goal of reducing the risk of human casualties.
Tools Used
R
Category
Predictive Analytics
Date
November 20, 2022
Challenge
This project presents a binary classification problem that seeks to identify and prevent tragic outcomes in mass mobilization protests.
Solution
Employed a range of techniques including exploratory data analysis, correlation analysis, and classification machine learning models. These efforts yielded an ROC AUC score of 86%, demonstrating the effectiveness of the approach.
