NYC School of Data 2025

Join machine learning engineer Isaac Donis and designer/UX expert Tim Harris for an illuminating data-driven presentation on New York City's congestion pricing through the lens of NYC Department of Transportation cameras. Born as a passion project, they've developed a custom computer vision approach to capture and analyze traffic patterns across the city, creating a unique grassroots dataset that offers fresh insights into the real-world effects of congestion pricing on our streets.

This session will showcase their methodology for counting vehicles per minute across DOT camera feeds and present their findings in accessible, visually compelling formats. Perfect for data enthusiasts, policy-minded New Yorkers, transportation advocates, and anyone curious about how technology can provide new perspectives on urban challenges. The presentation will conclude with a Q&A session where audience members can engage directly with the presenters about their techniques, findings, and implications for the future of NYC transportation.

OSCUR brings together data science (visualization, data management, analytics, modeling, HCI), policy, and engineering (sustainable & resilient urban infrastructure, network science) to tackle important scientific questions and help urban data analytics efforts around the world.  We will discuss the platform and demo two example applications.
UTK: Using the Urban Toolkit (UTK) we will provide and interactive tutorial demonstration that shows how eight different datasets from NYC Open Data portal  (including crime and noise reports; restaurants, parks, and subway locations; sky exposure; school quality reports and taxi pickups) can be used to create an integrated map-based visualization to help us to better understand the uniqueness of different neighborhoods. We will provide preprocessed datasets and shapefiles, and walk participants through the creation of "knots" that attach data from each of the loaded datasets to a neighborhood in the interactive map. (https://urbantk.org/utk/)

Curio: Using Curio we will demonstrate how to conduct a micro-scale environmental analysis, using Milan as a test case. We will first show how weather data are loaded into Curio and create a spatial join to link the data to the neighborhoods of Milan. We will then show how to calculate the Universal Thermal Climate Index (UTCI), and create linked visualizations to highlight neighborhoods with high UTCI and a large population of older adults. This example will be used to show how the Curio framework supports collaborative urban visual analytics more generally. (https://urbantk.org/curio/)

Other OSCUR related projects include:
· Project Sidewalk (https://makeabilitylab.cs.washington.edu/project/sidewalk/) 
Our NYC research has been featured in NY Times:
· Mapping the Shadows of New York City: Every Building, Every Block (https://www.nytimes.com/interactive/2016/12/21/upshot/Mapping-the-Shadows-of-New-York-City.html)
· The Coronavirus Quieted City Noise. Listen to What’s Left. (https://www.nytimes.com/interactive/2020/05/22/upshot/coronavirus-quiet-city-noise.html)

In July 2024, the NYC Department of Sanitation added several new datasets to Open data that help tell the story of where garbage goes. These datasets provide data on the annual tons of refuse sent to each of the transfer stations used by DSNY, as well as details on the landfills and WTE combustors that receive those tons go after they leave the transfer station. The data presented in three tables that can be linked to show the quantities of NYC trash that flow through transfer stations and on to different disposal locations. The tables also show information on costs, travel distances, and modes of transportation.

In conjunction with two additional datasets already existing on Open Data, it is possible to estimate the flow of refuse from an individual district through the transfer station and on to specific disposal sites outside the city. This is the closest New Yorkers have ever come to being able to follow their trash on an ongoing basis. Knowing this information puts fresh perspective on what it means to dispose of your trash; and how doing other things with it -- like recycling, composting, donating, reusing and reducing -- is a better option.

This presentation will tell the story of how these datasets were requested and ultimately put up on NYC Open Data, and will provide a "how to" primer on using and interpreting the information in maps and nonspatial analysis. This information is especially relevant to residents of disadvantaged communities in NYC where waste transfer stations cluster, and to activists in community composting, urban agriculture, and local reuse enterprise who see NYC's zero waste future as one in which materials are kept local to the extent possible, and risks are not exported to other communities.

Join software engineer Yufeng Zhao for a session combining technical insights with hands-on practice in creating searchable collections of urban text. Yufeng will first share how they built "all text in nyc" (https://alltext.nyc) - a search engine that lets you find any text visible in New York City's streetscape, from shop signs to graffiti.

The session is split into two parts: First, there will be a demonstration of how to collect and process panoramic images, implemented OCR on curved text, and created a geographic search interface. Then, we'll roll up our sleeves for a hands-on workshop where you'll use an open-source library to create your own small collection of searchable street text.

This session is perfect for civic technologists, urban researchers, and anyone interested in discovering new ways to document and analyze our city's textual landscape. No advanced programming experience required - if you can run a Python script, you're ready to participate!