Emptybox Journal

# Urban Science Intensive

1/30/2021

# Projects

# A Finance Map of NYC: Supporting PACE Financing to Drive Large-Scale Energy Efficiency Retrofits

Sponsor: NYU Stern Center for Sustainable Business

Project Abstract

Building a database for Local law 97 to be queried by users

# The Impact of Urban Agriculture on Food Chain Resiliency and Food Equity in New York City

Sponsor: NYU Stern Center for Sustainable Business

Project Abstract

Mapping NYC food resiliency identify urban farming hotspots, distribution gaps, and neighborhood impacts, contributing to NYC’s food resiliency plans.

# Predicting and Preventing Lead Paint Poisoning in New York City Among Young Children

Sponsor: Research & Analytics / Office of the NY Attorney General

Project Abstract

New York City Lead Poisoning Prevention Law (known as Local Law 1) requires landlords of residential rental buildings to identify and fix lead paint hazards in the apartments with children under the age of six. Local Law 1 also requires the Department of Housing Preservation and Development (HPD) to carry out inspections any time a child tests with a lead level above five micrograms.

# Predictive Modeling of Opioid Overdose Risk for Targeted Public Health Interventions

Sponsor: Machine Learning for Good Laboratory

Project Abstract

We are currently working with the Rhode Island Department of Health (RIDOH) on the NIH-funded PROVIDENT project. PROVIDENT has two goals: predictive modeling of opioid overdose risk, for geographic targeting of public health interventions by RIDOH, and incorporating these models into a randomized controlled trial to evaluate the impact of targeted interventions on reducing overdose fatalities. The capstone project team will contribute to this ongoing research in two ways: (1) comparative evaluation of machine learning and deep learning models for opioid overdose prediction, and (2) modeling subpopulation-level heterogeneity in opioid overdoses.

# Notes from webinar

  • has datasets for target variable
    • ODs and non-lethal ODs
  • EMS data
  • census block level
  • data scarcity is a challenge
  • how do you accurately target interventions for communities facing OD risk?

# Mapping Construction+Demolition Waste Flows for Recovered C+DW Use in City’s Capital Program

Sponsor: Town+Gown:NYC and CDW Working Group

Project Abstract

New York State regulates CDW—its generation, recycling and reuse—and collects all data on CDW from private waste haulers and transfer stations/recycling facilities. There is no city source of data for CDW. For the city to innovate policy with respect to CDW by leveraging its capital program to close material loops, generating environmental sustainability and financial sustainability benefits, it is important to understand where CDW goes after the demolition process through the recycling process.

# Applications of Machine Learning and Remote Sensing for Assessing Open Street Map Completeness on a Global Scale

Sponsor: New Light Technologies, Inc.

Project Abstract

Open Street Map (OSM) is a well-known platform that leverages crowd sourced knowledge as ground truth for asset mapping; however, there are large gaps in coverage and quality in many parts of the world, particularly in developing countries. This project will identify suitable remote sensing data sources to drive a supervised learning algorithm trained on areas of OSM that have been mapped with high confidence to quantify a scale of OSM completeness for “unseen” (untrained) areas. The desired output will be a proofof-concept methodology for assessing OSM completeness on a global scale and a supplemental interface for visualizing results

# FloodSense - Computer Vision for Urban Street Flood Detection

Sponsor: CUSP / The Floodsense Project

Project Abstract

In NYC, sea level rise has led to a dramatic increase in flood risk, particularly in low-lying and coastal neighborhoods. Urban flood water can impede pedestrian and vehicle mobility, and also can contain a diverse array of contaminants, including fuels, raw sewage, and industrial/household chemicals. For this capstone project, the team will train, test and deploy computer vision (CV) and deep learning (DL) models for the detection of street flood events. Existing labeled datasets will be used for training. In addition, an unlabeled NYC street image dataset will be provided for labeling and training of a NYC centric model.

# Examining Body Language in Road-Crossing Behavior

Sponsor: CUSP SimSpace

Project Abstract

Road-crossing behavior among pedestrians is a necessary but often risky endeavor in many cities

# Center for Opioid Epidemiology & Policy Dashboard

Sponsor: NYU Center for Opioid Epidemiology & Policy

Project Abstract

Deaths from opioid overdose in the United States increased by nearly 400% between 2000-2018. State and local governments have enacted various opioid-related policies in an effort to mitigate harms from the opioid crisis, resulting in a policy landscape that is both complex and dynamic. There is a need for rigorous opioid-policy evaluation studies, as policymakers and stakeholders seek to understand the most effective policies. A comprehensive, merged dashboard of opioid morbidity and mortality, state- and local-level opioid-related laws, and local information on enforcement of such laws would be a powerful and informative tool to synthesize evidence on this complicated topic.

# California Dream Index

Sponsor: California Forward (CA FWD)

Project Abstract

The California Dream Index (“CDI”) provides a measure of California’s economic mobility. Measuring 10 trackable indicators for economic mobility, security and inclusion through one powerful platform, data can be analyzed by race and ethnicity region, and county, providing the ability to do “apples to apples” regional comparisons across the state. This project will build on the version 1.0 version of the tool by extending the CDI from a county and regional scale to provide more geographically and demographically granular profiles. Those will include city and neighborhood scale profiles as well as statistical composites of subgroups embodying the state of the California Dream.

# Measuring Spatial Behavior Among Pedestrians During Pandemics

Sponsor: CUSP SimSpace Lab

Project Abstract

The goal of this work will be to develop schemes for analyzing video data for signatures and motifs of human spatial behavior in busy street scenes.

# Modality-Agnostic Road Traffic Monitoring

Sponsor: NYU Music and Audio Research Laboratory

Project Abstract: Monitoring road traffic is key to ensuring user safety and smooth operation. Increasing traffic volumes impact the stress level of commuters and increase noise levels in communities, leading to health problems. Local authorities need reliable monitoring systems to create policies to help mitigate this. Ideally, automatic monitoring systems should be able not only to count vehicles but also to detect the type of vehicle (e.g. car, truck). In this project we aim to develop a system for classification of vehicles that is able to use audio or video instinctively, complying with different privacy regulations and budget constraints

# New York State’s Broadband Coverage Mapping for the Reimagine New York Commission

Sponsor: Schmidt Futures, on behalf of the Reimagine New York Commission

Project Abstract: The Reimagine New York Commission has ratified an initiative to establish an accurate baseline of consumer broadband experience in New York State. This initiative can position New York as a national leader in broadband data collection, provide information for use by any community for broadband expansion projects, and support pursuit of grant funding available from the FCC, USDA, and future state programs.

In this Capstone Project we will aggregate different datasets in order to build an accurate, lotlevel coverage map on wired broadband availability, costs, geographical gaps, and competition, to better inform future deployment efforts.

# Do Broken Windows Encourage Criminality?

Sponsor: Dynamical Systems Laboratory

Project Abstract

Criminology theories are at the core of effective crime prevention strategies, which are critical tools in thriving urban areas. One of the most famous and controversial criminology theory is the so-called broken windows theory, stating that episodes of public incivility and disorder foster serious crime. While this hypothesis informed policy-making in several urban areas, including New York City, its validity and implications on law enforcement are still subject of debate. In this project, we propose a mathematically-principled analysis of this theory, aiming at unraveling cause-effect relationships between public incivility, crime rates, and enforcement activities

# US Ignite, Fort Carson and the City of Colorado Springs : Sensor Network and Digital Dashboard

Sponsor: US Ignite(opens new window)

Project Abstract

The Fort Carson Army Garrison has a close relationship with Colorado Springs due to the nearly 30,000 soldiers and staff members that commute between the city and the installation on a daily basis. This traffic volume alone creates congestion at the entry gates and surrounding roadways during peak travel times. When coupled with the extreme weather conditions of the region, decision makers at both the city and installation level are continuously being challenged on how to manage operations. By leveraging the available data resources and supplementing the region with a deployment of smart sensors this project aims to bring critical information to these organizations in a visual format that is both intuitive and representative of real world conditions. Utilizing the latest in advanced wireless technology ensures the data is accurate and dependable.

# Urban Inventories

Sponsor: Urban Modeling Group(opens new window)

Project Abstract

Cities are spectacularly complicated places. Even a sidewalk is not a simple, uniform and paving strip. Sidewalks are littered with driveways, cut curbs, steps, grates, fire hydrants, utility poles, benches, and trees, and much more. While documentation exists for some items, the information is often incomplete or has only general geo-referencing. Without a relatively comprehensive inventory and representation of the built environment, the concept of an urban digital has little heft. Thus, this capstone investigates how various forms of machine learning can be brought to bear to achieve automatic object identification from high density aerial laser scanning data sets.

# Internet-of-Things Security and Privacy

Sponsor: NYU Tandon MLab(opens new window)

Project Abstract: Smart home IoT (Internet-of-Things) devices are gaining popularity in average consumer homes. These “smart” devices, such as cameras, plugs, TVs, dishwashers, etc, are also known to pose various security and privacy threats (e.g., your Alexa listening to you), but the opaque nature of these devices makes it difficult to discover security and privacy vulnerabilities. In this project, we plan to systematically discover Internet-of-Things (IoT) security and privacy issues from a data-driven perspective and develop mitigation from both a technical and policy perspective. We will build upon existing work from IoT Inspector (https://iotinspector.org/), analyze existing datasets, and/or extend IoT Inspector to gather more data-driven insights. At the end of this project, we will provide the public with either an improve tool or report to keep consumers informed and protected.

# Open Police Data: Collection, Analysis, and Outcomes

Sponsor: What Works Cities(opens new window)

Project Abstract: Many cities have open data portals and are committed to transparency and sharing data with their residents, but this commitment sometimes fails to extend to the police department leading towards delegitimization and distrust. What is the current landscape of open police data in US cities and what variables lead to some local police departments to be more transparent than others? This project involves building a web-scraping application for data collection, qualitative data collection through interviews, and analyzing and reporting out on the collected data and findings.

← Innovative Cities →