Tandon Team Takes Home Top Honors at Association for Computing Machinery Conference

The Association for Computing Machinery (ACM) Special Interest Group on Management of Data (SIGMOD) holds an annual conference universally considered to be the world’s top forum for database researchers, practitioners, developers, and users. There they can explore cutting-edge ideas and results; exchange techniques, tools, and experiences; and network with leading figures in the field.

This year’s event, held in Houston, proved to be a banner one for members of NYU Tandon’s Visualization and Data Analytics (VIDA) Research Center, who are engaged in applying the latest advances in computing to problems of critical societal importance and in generating the technological solutions that new data sources and data types demand.

A VIDA team answered conference organizers’ call for demo papers, and their efforts were rewarded with the top prize in that category. (Unlike a regular research paper, which simply details the scientific contribution made by its authors, a demonstration paper provides the chance to actually try out a working tool or data set.)

Co-authored by Research Assistant Professor Harish Doraiswamy, Professors Juliana Freire and Claudio Silva, and post-doctoral researcher Fabio Miranda (along with colleagues from École Polytechnique Fédérale de Lausanne and Brazil’s Universidade Federal Fluminense), “Interactive Visual Exploration of Spatio-Temporal Urban Data Sets using Urbane” describes a novel 3D visual analytics framework that supports data-driven decision-making in the design of new urban developments. 

 “Why do two regions in a city feel similar? Or different? What are the characteristics that determine the quality of a city?” the authors asked. While the architects, urban planners, and other policymakers who shape the future of the world’s cities typically rely on their own experience, set precedent, and data analyzed in isolation to answer those questions, the team asserts that the recent proliferation of spatiotemporal data sets from urban environments and social sensors provides another pathway, creating new and better opportunities for data-driven approaches. 

“Architects, for example, need to have a strong understanding of a neighborhood’s characteristics to identify potential sites for development,” the authors explained. “By using the available open data sets and comparing the neighborhood of interest with other neighborhoods, they can understand its strengths and weaknesses and establish performance thresholds from other well-known and well-performing neighborhoods. This will eventually facilitate the negotiation process with the city planner, who is concerned with maintaining the quality of the neighborhood.” To that end, the VIDA researchers developed Urbane, a platform that allows users to visualize a data set of interest at different resolutions over varying time periods.

At the ACM’s 2018 International Conference on the Management of Data, attendees were given a chance to see firsthand how Urbane could enable decision-makers to harness the power of big data to create vibrant, livable, sustainable communities. In addition to viewing two existing case studies (one involving NYC’s Financial District and the other identifying sites with development potential), visitors could choose their own data sets to examine, zoom into regions of interest, and even explore at the level of individual buildings. “To enable interactive exploration and analysis, the queries generated by the user interface have to be executed in real time – typically they should take less than a second,” Doraiswamy explains. “Thus, one of the main goals of the demo was to show how we could accomplish this.”

With troves of data being generated minute by minute in New York and other urban areas — from taxi rides and mass-transit status to noise complaints and incidents of crime — the VIDA team is helping stakeholders make sense of it all — and impressing fellow members of the ACM SIGMOD at the same time.