Domestic Violence Prevention Analysis
Predictive Modeling for Repeat Victimization Risk Assessment
Project Overview
This capstone project leverages data from the Chicago Women's Health Risk Study to develop a predictive model for assessing the probability of repeat victimization in domestic abuse cases.
The model provides crucial insights that can help social workers, law enforcement, and support organizations make more informed decisions when allocating resources and providing protection to victims.
Problem Statement
Domestic violence affects millions of people worldwide, with repeat victimization being a critical concern for victim safety and resource allocation.
Traditional assessment methods often rely on subjective judgment. This project aims to provide a data-driven approach to risk assessment, helping identify victims at higher risk of repeat abuse.
Technical Implementation
The project utilizes advanced machine learning techniques to analyze patterns in the Chicago Women's Health Risk Study dataset, implementing feature engineering and model validation to ensure reliable predictions.
Data Source
The analysis is based on the comprehensive Chicago Women's Health Risk Study, a longitudinal study examining factors associated with intimate partner violence.
This dataset provides rich information about victim demographics, relationship characteristics, and incident details that inform our predictive model.
Impact & Applications
The web-based tool enables:
• Social workers to assess risk factors quickly and objectively
• Law enforcement to prioritize cases requiring immediate attention
• Support organizations to allocate resources more effectively
• Researchers to identify key patterns in domestic violence cases
Key Features & Achievements
- Developed predictive model using machine learning algorithms to assess repeat victimization risk
- Created interactive web interface for real-time risk assessment
- Analyzed complex social and demographic factors contributing to domestic violence patterns
- Implemented statistical validation techniques to ensure model reliability
- Designed user-friendly tool for non-technical stakeholders in social services
- Collaborated with interdisciplinary team to address real-world social challenges
Repository Links
Main Project Repository: domestic-violence
Web Interface Repository: ddfloww-site
Original Study: Chicago Women's Health Risk Study
Live Application: Risk Assessment Tool