View Proposal
-
Proposer
-
Soo Huei Ching
-
Title
-
A Machine Learning Approach to Analysing Socioeconomic Determinants of Crime Rates in the United States
-
Goal
-
To examine how socioeconomic factors affect crime rates across U.S. states using statistical and machine learning techniques.
-
Description
- Crime remains a critical social issue in the United States, where rates vary significantly across regions and over time. Both violent and property crimes are shaped by complex and interrelated socioeconomic factors. Understanding these influences is essential for designing effective interventions.
This project aims to analyse the impact of socioeconomic determinants—such as education, income, employment, housing, and demographics—on crime rates across U.S. states. Two primary datasets will be integrated: the CORGIS Dataset Project, which provides state-level annual crime statistics, and the U.S. Census Bureau’s socioeconomic data. These sources will be merged by state and year to form a unified dataset for analysis.
Following data preparation, models such as Generalized Linear Models, Random Forest, and other machine learning algorithms will be applied to predict crime rates. The performance of these models will be compared using evaluation metrics such as Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R². The study will cover data collection, cleaning, exploration, modeling, and evaluation.
By uncovering patterns between socioeconomic conditions and crime, this project seeks to provide insights that can inform policymakers and community leaders in developing evidence-based strategies to reduce crime and improve social well-being.
- Resources
-
1. Goin, D. E., Rudolph, K. E., & Ahern, J. (2018). Predictors of firearm violence in urban communities: A machine-learning approach. PLoS ONE, 13(10), e0205151. https://doi.org/10.1371/journal.pone.0205151
2. Gesin, J. (2014). Socioeconomic determinants of violent crime rates in the U.S. Empirical Economic Bulletin, 7(Spring). Bryant University. https://digitalcommons.bryant.edu/ebook/23/
Data Sources
3. Whitcomb, R., Choi, J. M., & Guan, B. (n.d.). State crime CSV file. CORGIS Dataset Project. https://corgis-edu.github.io/corgis/csv/state_crime/
4. United States Census Bureau. (n.d.). American Community Survey tables. https://data.census.gov/table
-
Background
-
-
Url
-
-
Difficulty Level
-
Moderate
-
Ethical Approval
-
None
-
Number Of Students
-
1
-
Supervisor
-
Soo Huei Ching
-
Keywords
-
machine learning, socioeconomic factors, crime analysis
-
Degrees
-
Bachelor of Science in Statistical Data Science