A data visualization and analysis dashboard built in Kaggle to explore public safety incidents reported to Seattle police. This project leverages real-world data to uncover trends, analyze call types, and display incident locations across the city.
- Source: Seattle Open Data Portal
- Records: ~99,000+ incidents
- Date Range: May 15, 2015 – April 12, 2025
- Columns: 25 attributes including:
Reported Date,Call Type,Disposition,Precinct,Sector,Beat- Officer demographics and classification of incidents
Pandas– data manipulationMatplotlib– static chart visualizationPlotly– interactive charting (converted to matplotlib for compatibility)Folium– interactive mapping of recent incident locationsNumPy– numerical operations
- Line and bar plots of daily and weekly incidents
- Stacked visualizations by Precinct and Sector
- Top 5 call types by count and precinct
- Bar charts colored by trend (increase or decrease)
- Calculated recent 3-day average vs. 30-day average
- Colored bar plot to show rising or falling sectors
- The latest 100 incidents mapped using Folium
- Dispositions shown as popups
Each police beat was manually mapped with latitude and longitude values for accurate visualization. These coordinates were cleaned and merged into the dataset to support spatial plots.
- Incidents Per Day:
Line graphof counts over time - Precinct Breakdown:
Stacked barfor the last 30 days - Disposition Types: Horizontal bar chart
- Sector Trends: 3-day vs. 30-day trend comparison
- Call Type Frequency: Top 5 visualized in grouped bars
This dashboard is hosted and runs on Kaggle. To replicate or explore:
- Download the dataset
- Open in a Kaggle notebook or local Jupyter environment
- Run through the cells sequentially to view insights
- Add interactive filters by time, sector, and disposition
- Incorporate severity or urgency scoring
- Enable time-lapse animations using Plotly or Kepler.gl
- Deploy on Streamlit or Flask for public use
Created by: Damarcus Thomas
Kaggle Notebook: [Link to your Kaggle notebook]
Email: 146stat@gmail.com
This project demonstrates how public datasets can be used to surface critical insights around mental health, policing, and resource allocation across urban environments.