NYC Downtown at sunset


  • Assistant Professor Elizabeth Hénaff was recently selected for The Marron Institute’s Second Annual Seed Grant Award for her proposal on “The Impact of Flooding on the Urban Microbiome and City Residents’ Exposure to Sewage Pathogens.” The research team, which also includes Professors Andrea Silverman and Tega Brain (NYU Tandon), will evaluate how urban flooding changes the microbial community of urban surfaces, investigate the ways in which the resulting microbial fingerprint evolves over time after a flood event, and determine how long it takes for the community to return to a pre-flood profile.
  • Assistant Professor Elizabeth Hénaff’s research is also part of a new art exhibit at the Detroit Science Gallery. SCOPE: Theater of Collaborative Survival is centered on a set of aquariums, each containing living microenvironments sampled from the Gowanus Canal. The installation explores the multiple facets of the complex relationship we have with contaminated waterways.
  • Associate Professor Constantine E. Kontokosta, PhD, PE, has been selected as a Google AI Impact Grantee. Prof. Kontokosta and his research team will receive a $500,000 grant to partner with the New York City Fire Department (FDNY) analytics team to utilize artificial intelligence and machine learning models to understand and improve emergency response times for the 1.5 million emergencies FDNY answers each year, accounting for factors such as weather, traffic, location, type of emergency, and more.
  • Associate Professor Constantine Kontokosta also announced the launch of the NYC Energy & Water Performance Map, which was developed in partnership with the Mayor’s Office of Sustainability. Covering six years of data, the NYC Energy & Water Performance Map provides an interactive data analysis and query platform to better understand the energy and water efficiency of more than 20,000 of the largest buildings across New York’s five boroughs. This visualization tool was created by the NYU Urban Intelligence Lab research team members and graduates of NYU CUSP: Bartosz Bonczak, Cyrus Blankinship, Unisse Chua and Ian Stuart, led by Dr. Constantine E. Kontokosta, Associate Professor of Urban Science and Planning and Director of the Civic Analytics Program at NYU Marron Institute of Urban Management.
  • The Urban Modeling Group has completed the latest iteration of their flyover and are now just waiting for data. A huge thanks to all the students, staff, and collaborators who helped us make this very large deployment happen!
  • Postdoctoral Researcher Vincent Lostanlen wrote a chapter for a new book, “Florian Hecker: Halluzination, Perspektive, Synthese.” The book, published by Sternberg Press, presents essays by curators, researchers, theorists, and art historians on composer Florian Hecker’s work and its relation to topics ranging from musique concrète, Mallarmé’s poem “The Afternoon of a Faun,” and computer music.
  • Professor Julia Lane’s textbook “Big Data and Social Science: A Practical Guide to Methods and Tools” is listed at #13 of BookAuthority’s 100 best data science books of all time. The text shows how to apply data science to real-world problems in both research and practice, and provides practical guidance on combining methods and tools from computer science, statistics, and social science.


  • Mark Cartwright, Graham Dove, Ana Elisa Méndez Méndez, Juan P. Bello and Oded Nov.Crowdsourcing multi-label audio annotation tasks with citizen scientists.” Proceedings of the ACM 2019 CHI Conference on Human Factors in Computing Systems.
    • Annotating rich audio data is an essential aspect of training and evaluating machine listening systems. We approach this task in the context of temporally-complex urban soundscapes, which require multiple labels to identify overlapping sound sources. Typically, this work is crowdsourced, and previous studies have shown that workers can quickly label audio with binary annotation for single classes. However, this approach can be difficult to scale when multiple passes with  different  focus classes are required to annotate data with multiple labels. In citizen science, where tasks are often image-based, annotation efforts typically label multiple classes simultaneously in a single pass. This paper describes our data collection on the Zooniverse citizen science platform, comparing the efficiencies of different audio annotation strategies. We compared multiple-pass binary annotation, single-pass multi-label annotation, and a hybrid approach: hierarchical multi-pass multi-label annotation. We discuss our findings, which support using multi-label annotation, with reference to volunteer citizen scientists’ motivations.


  • A paper by two CUSP graduate students, Yusu Qian and Urwa Muaz, was accepted by the 2019 ACL Student Research Workshop (SRW), which will be held in Florence, Italy by the end of July! The paper, Reducing Gender Bias in Word-Level Language Models with A Gender-Equalizing Loss Function, proposes a novel method to address gender bias in the neural language models. The authors introduce a penalty term to the objective function of the language model to penalize the discrimination against gender.
  • Graduate student Yusu Qian will also present the proposal Gender Stereotypes Differ between Male and Female Writings at the 2019 ACL Student Research Workshop.
  • Graduate student Sarah Sachs recently published a new piece on data.tale() with her team from the recent CUSP Data Dive event in London. Emergency Deployment Optimized for a Dynamic City explores their task to understand London’s traffic patterns in order to optimize the deployment of ambulances and minimize the run-time required for ambulances to reach an incident.