metadata

title: README
emoji: 🌍
colorFrom: blue
colorTo: green
sdk: static
pinned: true
license: mit

OSCUR: Open-Source Cyberinfrastructure for Urban Computing

OSCUR is transforming urban computing by making complex city data accessible and actionable through open-source tools. Funded by a $5M National Science Foundation grant, we empower researchers, city planners, and communities to tackle urban challenges like transportation, climate change, and environmental justice.

Vision

We Aim to Revolutionize Urban Computing

Foster Partnerships: Advance open and reproducible urban analytics through collaboration.
Leverage NSF Initiatives: Harness the strengths of existing NSF-funded projects.
Promote Civic & Global Programs: Support civic initiatives, research, and education worldwide.
Democratize Urban Computing: Make analytics accessible to all globally.

Outcomes

A novel open-source cyberinfrastructure for reproducible urban computing.
A cohesive and engaged urban computing community.
A path toward healthier, more resilient, accessible, and sustainable cities.

They Say It Best...

"While open data is increasingly available across a host of urban areas - from resource consumption to housing and infrastructure — its potential remains largely untapped because of unique challenges related to the diversity and scale of the data and the complex computations required to obtain trustworthy insights. OSCUR intends to unlock that potential"
— Claudio Silva, NYU Tandon School of Engineering

"With OSCUR, we're trying to democratize urban computing and empower a broad range of stakeholders to analyze urban data at scale. We want to remove the silos in which urban data typically lives, lower the barrier to entry for urban computing and facilitate better collaboration. By providing a unified platform with powerful tools, we hope to accelerate research and development in this critical field."
— Juliana Freire, NYU Tandon School of Engineering

Show more quotes

"OSCUR empowers us to turn data into action for healthier cities. One of the key aspects is lowering the barriers to accessing this wealth of technologies we have been building over the past years. We’re creating analyses and visualizations that ensure that community members are aware of how they are impacted."
— Fabio Miranda, University of Illinois Chicago

"Many cities and governments and industries are putting data out there, but it’s just very hard to use. Imagine if you hire the best chef in the world and give them the best ingredients, but you don’t give them any tools. No matter how good the chef is, if they don’t have the proper tools, they can’t make a good dish."
— Sybil Derrible, University of Illinois Chicago

"The true beauty and promise of OSCUR is in how it attempts to unify long-standing and deeply interconnected problems in urban science that often have disparate approaches spread across disciplines."
— Jon E. Froehlich, University of Washington

"Because of their history, relationships between government and marginalized communities can be fraught with distrust — including distrust towards the data that government releases. You need to put time and effort into building something that is trustworthy and valuable for the community, in terms of allowing them to analyze the data in a way that matters to them."
— Federica Fusi, University of Illinois Chicago

Sources: Urban computing gets a boost — NYU Tandon | UIC researchers join national project — UIC Today

Key Features

Unlock the Power of Urban Data

Explore OSCUR’s potential at oscur.org and join the urban computing community!

Datasets

Through all research outputs by OSCUR, datasets will be used, while as much as possible striving for open-data. Therefore, to make it easier, we host below a list of open-datasets for OSCUR users to more rapidly access and analyse via the Hugging Face Datasets library. To name a few, this will be including urban tabular datasets, annotated visual data, sensor readings, human behavior markers, and other structured data tailored for urban computing scenarios. Each dataset is formatted for direct integration into machine learning workflows and visual analytics pipelines, supporting applied research and education.

Note: Access our curated datasets with ease! For example, to load NYC Taxi Vis is the taxisvis1M dataset, use the following code:

from datasets import load_dataset dataset = load_dataset("OSCUR/taxisvis1M")

Simply replace taxisvis1M with other dataset names to explore additional datasets. Here are some examples:

NYC Taxi –– taxisvis1M : Public New York City taxi pickup and drop-off datasets.

NYC Service Requests –– NYC_311 : 311 Service on-street Requests from 2010 to present in New York City datasets.

NYC Buildings Information –– pluto : The Primary Land Use Tax Lot Output (PLUTO™) containing extensive land use and geographic data at the tax lot level in NYC.

Libraries in the OSCUR Ecosystem

Explore the tools powering OSCUR:

Data Collection & Discovery

Scrape and collect urban data from diverse sources.

CSAI Data Gathering – Generate and enhance geospatial datasets for the United States Department of Transportation (USDOT) Complete Streets AI Initiative, covering Infrastructure, Traveler Behavior/Safety, and Context.

Curation & Transformation

Pre Process & Transform machine-learning-ready urban datasets.

🍳 No libraries yet, but we’re cooking something great!

Analytics & Modelling

Perform advanced analytics and machine learning on urban datasets.

UrbanMapper – Augment & Enrich plain urban layers using urban datasets' basic to ML-powered and LLM- statistics and aggregations using a Scikit-Learn-like urban pipeline!

Visualization

Create interactive visualizations for urban insights.

🍳 No libraries yet, but we’re cooking something great!

Let’s build smarter cities together! 🌟

Cheers,

@OSCUR team