What's Happening at CUSP: December 2017

People

Welcome to Ray Mohabir, a Tandon Undergrad who is working as a Digital Video Production Assistant with Debra Laefer. He is producing a digital video fly through of the O’Connell Street area of the 2015 LiDAR Dataset.

Welcome to Sabrina Hadded, who has joined the Data Facility as a part-time Research Project Manager.

Congratulations to Debora Laefer, who has been appointed to the Editorial Board of ISPRS Journal of Photogrammetry and Remote Sensing, the top journal in remote sensing.

Congratulations to Joe Chow who has been awarded the Council of University Transportation Centers’ 2017 Cambridge Systematics New Faculty Award. The award, which recognizes outstanding teaching and research contributions to the transportation field, will be given to him at the annual banquet in Washington DC on January 6th, 2018.

Congratulations to Constantine Kontokosta, who has received funding from the Trust for Governor’s Island to support a postdoc for a new QC on Governor’s Island. They will be deploying sensors around the island and working on mobility analyses. He has also been invited to serve on the organizing committee of the National Academy of Sciences’ Data and Computation for Urban and Community Sustainability

Constantine Kontokosta and his group – Yuan Lai, Bartosz Bonczak, Sokratis Papadopoulos, Bo Hong, Awais Malik, and Nick Johnson – won the Best Data Visualization Award in the UN Data for Climate Action challenge for our work in developing a dynamic spatio-temporal model of urban GHG emissions integrating statistical and physical models derived from data from Waze, Earth Networks, Plume Labs, Weather Sentry, NYC Mayor’s Office, DOT, and DOE, among others. They beat out an initial group of 450 teams from 67 countries and then a semi-finalist group of 97 teams from universities around the world.

Bartosz, Sokratis, and Constantine partnered with the Urban Green Council to produce the 5^th NYC Energy and Water Use Report released by the Mayor’s Office of Sustainability. Current and prior year reports and data can be found here.

Kontokosta, Constantine E., Mitchell Weiss, Christine Snively, and Sarah Gulick. “NYC311.” Harvard Business School Case 818-056, October 2017. (Revised November 2017.). The work focused on detecting and quantifying bias in 311 citizen complaint data.

Kontokosta, Constantine E., and Awais Malik. “The Resilience to Emergencies and Disasters Index: Applying big data to benchmark and validate neighborhood resilience capacity.” Sustainable Cities and Society36 (2018): 272-285. The paper presents a new method to quantify local resilience capacity and presents an approach to validation using 311 as a measure of community recovery following an event.

Kontokosta, Constantine, Boyeong Hong, Awais Malik, Ira M. Bellach, Xueqi Huang, Kristi Korsberg, Dara Perl, and Avikal Somvanshi. “Predictors of Re-admission for Homeless Families in New York City: The Case of the Win Shelter Network.” accepted to the Journal of Technology in Human Services.

Sayarshad, Hamid R., and Joseph YJ Chow. “Non-myopic relocation of idle mobility-on-demand vehicles as a dynamic location-allocation-queueing problem.” Transportation Research Part E: Logistics and Transportation Review106 (2017): 60-77. Their algorithm repositions idle vehicles in anticipation of future demand and we used taxi data in NYC to test it.

Huy Vo and Tim Savage delivered a poster and paper, Deep Learning and Finance: Here We Go Again, at the Inaugural Tech Conference of the National Association of Business Economists. They show that deep learning is no better using two workhorse models in empirical finance than much simpler approaches. But deep learning comes at considerable computational cost, even with the most sophisticated architectures available. Oren Etzioni mentioned the poster in his keynote talk saying that whether regression or DL, a human was still making the decision regarding the model and its interpretation.