This page explains how the site works, where the data comes from, and how to get the most out of filters, trends, and the statistical dashboard. It’s meant to guide both first-time visitors and students doing deeper analysis.
Start on the Home page to see what this project does. Then browse a few arrest records to get a feel for the fields and layout.
Use Filter & Search to narrow arrests by borough, precinct, offense, demographics, and date. This is the best place for focused analysis.
Visit the Statistical Dashboard and the Monthly / Weekly Trends pages to see the bigger picture over time.
Use this app as a learning dataset based on real NYC Open Data, sampled to keep queries fast and visualizations responsive.
High-level introduction and core feature cards (Browse Arrests, Filter & Search, Crime Ranking, Trends).
Paginated table of individual arrest records with details like date, borough, offense description, law category, age group, gender, race, and approximate location.
Advanced filters to narrow down arrests by borough, precinct, offense, law category (felony / misdemeanor / violation), age group, gender, and race. Use this when you want a very specific slice of the data.
Aggregates the data to show which offenses appear most frequently across NYC. Helpful for quickly spotting the top recurring charges.
Shows a summary view: total arrests, most common offense, top borough, approximate average age, and live charts for arrests by borough and monthly trends computed from the current dataset (not hard-coded numbers).
Visualizes total arrests over time using real arrest dates. Monthly trends show the broader pattern; weekly trends highlight shorter-term spikes and dips.
All records are pulled from the official NYC Open Data API for NYPD arrests. Our backend fetches the data in pages, cleans it, and stores a curated subset in our own MongoDB database for faster queries and visualizations.
The dataset reflects the latest available months of the current year from NYC Open Data. In our current setup, the records typically span from the past several months, depending on what the API has published when we seed the database.
The full NYC arrest dataset is very large. For this project, we fetch tens of thousands of rows from the API and then perform stratified sampling by borough (around 400 records per borough). This keeps the app responsive while still preserving realistic borough-level and temporal patterns.
The data can be refreshed any time by re-running the seeding script. Each run pulls the latest NYC Open Data records, re-samples them, and recalculates the monthly and weekly trends that power the charts.
When everything loads correctly from the database, the charts display labels like (Dataset) to indicate they are based on real aggregated values. If something goes wrong (for example, the DB is empty or an API call fails), the app falls back to a small built-in sample so the page still renders, labeled as (Sample).
A free account is required to access detailed analytics, demographic insights, and arrest details. Basic browsing of the arrest list is available without an account.
If a number or chart looks strange, it might be due to sampling, API limits, or how NYC reports certain categories. You can cross-check against the original NYC Open Data portal.