https://github.com/githubasr2001/austin_crimes
https://github.com/githubasr2001/austin_crimes
Last synced: about 2 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/githubasr2001/austin_crimes
- Owner: githubasr2001
- Created: 2025-03-03T19:46:07.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-03-03T19:53:56.000Z (3 months ago)
- Last Synced: 2025-03-03T20:34:43.000Z (3 months ago)
- Language: Python
- Size: 0 Bytes
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Austin Crime Data Analysis: Insights and Predictive Analytics
I'm excited to share my recent project analyzing Austin's crime data, where I built an interactive dashboard and predictive model to uncover patterns in criminal activity across the city.
## Project Overview
Using Python and a suite of data science tools, I developed a comprehensive crime analytics platform that transforms raw incident data into actionable intelligence. The dashboard provides law enforcement and community stakeholders with:
- Real-time visualization of crime patterns by location, time, and offense type
- Predictive modeling to forecast case clearance probability
- Spatial analysis to identify high-risk areas
- Temporal trend analysis showing crime fluctuations by hour, day, and month
## Key Findings
📊 **Temporal Patterns**: Identified peak crime hours and days, enabling optimized patrol scheduling
🗺️ **Spatial Distribution**: Mapped crime hotspots across council districts and location types, revealing significant disparities in criminal activity
🔍 **Offense Analysis**: Categorized and visualized predominant crime types, supporting targeted prevention strategies
⚖️ **Clearance Insights**: Built a Random Forest classifier achieving strong prediction accuracy for case resolution likelihood
## Technologies Used
- **Data Processing**: Pandas, NumPy, Scikit-learn
- **Machine Learning**: Random Forest Classifier with feature importance analysis
- **Visualization**: Matplotlib, Seaborn, Plotly Express
- **Interactive Dashboard**: Streamlit with custom CSS styling
- **Geospatial Analysis**: MapBox integration for crime mapping
## Impact
This project demonstrates how data science can support evidence-based policing strategies and community safety initiatives. The interactive dashboard allows stakeholders to:
- Filter data dynamically by date, offense type, district, and clearance status
- Generate downloadable reports for different analytical perspectives
- Identify key factors influencing crime clearance rates
- Visualize family violence incidents and distribution patterns
## Future Directions
I plan to enhance this project by incorporating demographic data, socioeconomic indicators, and additional time-series forecasting to provide deeper insights into crime causality and prevention opportunities.