This project focuses on visualizing motor vehicle collision data in New York City for the years 2021-2023. Utilizing the Altair visualization library in Python, the project aims to clean, summarize, and visually represent crash data in a user-friendly and insightful manner.
The dataset, sourced from the NYC Open Data portal, includes comprehensive details on motor vehicle collisions in NYC, covering various attributes such as date/time, borough, zip code, latitude/longitude, injuries/fatalities, contributing factors, and vehicle types.
- Time Series Analysis: Analyzes trends over time, including hours of the day, days of the week, and monthly patterns.
- Incident Attribute Counts: Aggregates data by various attributes such as injuries, boroughs, and vehicle types.
- Word Cloud: Visualizes common contributing factors to accidents.
- Borough Filtering: Allows users to filter data by borough, enabling localized analysis.
Feedback from non-expert users played a crucial role in refining the visualizations:
- Chart Labeling: Standardized axis labels across charts for uniformity and clarity.
- Heatmap Coloring: Improved the color scheme of the heatmap for better clarity and visual appeal.
- Word Cloud Density: Retained the word cloud in its existing form but educated users about its typical appearance and purpose.
- Filtering by Borough: Users appreciated the ability to filter data by borough, which enhanced localized data analysis.
- Tooltip Formatting: Adjusted tooltips for consistency and clearer formatting.
These refinements led to a more user-friendly and informative data visualization experience, enhancing both the visual appeal and the interpretability of the data.
To explore the visualizations, follow the link at https://visualization-final-ns5lsxqc2m5u9aiavocaym.streamlit.app/#nyc-motor-vehicle-collisions-visualizations
- NYC Open Data
- User feedback participants