DSI Seed Fund Past Projects

All Projects

• Psychology, Organizational Behavior and Neuroscience Literatures: Harnessing Data Science to Unify DEI Findings in Academic Literature and the Popular Press 2022

  Valerie Purdie-Greenaway, Psychology

  Alfredo Spagna, Psychology

  Peter Bearman, Sociology

  Jennifer Manly, Neurology

  Smaranda Muresan, DSI, Computer Science

   

  This team will develop a shared understanding of how diversity and inclusion (D&I) is conceptualized and studied in the academic literature and compare academic research on D&I to what is found in popular press outlets. The project will draw from social psychology, organizational behavior, and social-cognitive neuroscience to create a baseline for understanding the structure of scientific knowledge related to D&I and to understand what kinds of D&I research finds its way into the popular press.

• Using Data Science and Causal Inference to Estimate the Community-Level Impact of Police Behavior on Psychological Distress: The Case of No-Knock Search Warrants in Chicago 2022

  Gerard Torrats-Espinosa, Sociology

  Kara Rudolph, Public Health

   

  This team proposes to create a novel linkage of police administrative records that capture highly detailed information on all search warrants that the Chicago Police Department executed from 2012 to 2020. They will document spatial and temporal patterns of search warrant use across Chicago’s neighborhoods.

• Racial Inequality in Police Violence: Injuries and Fatalities from Police Use of Force 2022

  Jeffrey A. Fagan, Law, Public Health

  Rajiv Sethi, Barnard, Economics

  Elizabeth Ananat, Barnard, Economics

  Morgan C. Williams, Jr., Barnard, Economics

  Brendan O’Flaherty, Economics

  José Luis Montiel Olea, Economics

   

  This project will create a data archive on non-fatal injuries and fatalities from police encounters—data that may be harmonized and integrated with other increasingly detailed datasets on police killings—and provide estimates of a continuum of police use of force. The new database will provide capacity and research opportunities for departments, schools, laboratories, and students across the university on an urgent public policy issue.

• Positioning Energy Storage Technologies with Stochastic Climate Scenarios 2022

  Upmanu Lall, Earth and Environmental Engineering

  Bolun Xu, Earth and Environmental Engineering

   

  This project combines data-driven renewable energy simulations with model-based storage pricing models to quantify the financial value of various energy storage technologies in integrating renewables and mitigating climate change in a decarbonizing electric power system.

• Application of Gaussian Mixture Regression to Obtain Useful, Actionable Air Pollution Data from Consumer-Grade, Low-Cost Monitoring Devices 2022

  Xiaofan (Fred) Jiang, Electrical Engineering

  Daniel Westervelt, Lamont-Doherty Earth Observatory

   

  This team will develop and apply a novel, globally applicable, bias correction algorithm to a fast-growing global network of consumer grade, low-cost air quality sensors. This method will allow users to obtain high-quality data from raw, unvalidated sensor data, thereby empowering communities to better understand their air pollution exposure and take action.

• Using Machine Learning to Measure Racial/Ethnic Bias in Obstetric Settings 2022

  Veronica Barcelona, Nursing
  Kenrick Cato, Nursing
  Dena Goffman, Obstetrics and Gynecology
  Coretta Green, New York-Presbyterian
  Anita Holman, Obstetrics and Gynecology
  Janice James Aubey, Obstetrics and Gynecology
  Bernadette Khan, New York-Presbyterian
  Kenya Robinson, New York-Presbyterian
  Maxim Topaz, Nursing

   

  This team will examine the association between linguistic bias and pregnancy-related morbidity among birthing people from 2017-2019 at two hospitals. They will use natural language processing approaches to: 1) identify stigmatizing language in clinical notes, 2) examine patterns of language use by race and ethnicity, and 3) study associations between language use and pregnancy-related morbidity.

• ​​Randomized Methods for High-Throughput Characterization of Tokamak Sensor Streams 2021

  James Anderson, Electrical Engineering
  Michael Mauel, Applied Physics
  Jeffrey Levesque, Applied Physics

   

  Fusion science seeks to advance our fundamental understanding of physics and make plasma fusion viable for applications such as clean energy production. Tokamak fusion reactors generate vast and rich data sets obtained through multiple sensing modalities. The goal of this project is to develop new robust and efficient methods rooted in randomized numerical linear algebra for analyzing and characterizing complex fusion discharge dynamics.

• mHealth-Based Data Science Approach to Examine the Association of Minority Stressors with Sleep Health in Black and or Latinx Sexual and Gender Minority Adults 2021

  Billy Caceres, Nursing
  Ipek Ensari, Data Science Institute
  Kasey Jackman, Nursing

   

  This pilot study will use data science techniques to leverage ecological momentary assessment and consumer sleep technology to phenotype sleep health profiles in Black and Latinx sexual and gender minority adults. The investigators will use 30 days of daily electronic diaries and actigraphy to examine the associations of daily exposure to minority stressors (such as experiences of discrimination and anticipated discrimination) with sleep health among Black and Latinx sexual and gender minority adults.

• Harnessing Big Data and ML to Learn Optimal, Individualized Dynamic Treatment Rules to Prevent Opioid Use Disorder Relapse 2021

  Sean Luo, Psychiatry
  Min Qian, Biostatistics
  Kara Rudolph, Epidemiology

   

  Pharmacologic treatment of opioid use disorder (OUD) is complicated by the likely absence of a one-size-fits-all best approach; rather, “​optimal​”​ dose and dose adjustment are hypothesized to depend on person-level factors, including factors that change over time, reflecting how well the individual is responding to treatment. ​This team will use harmonized data from multiple existing clinical trials with natural variability in OUD medication dose adjustments over time to 1) learn optimal dosing strategies​,​ and 2) estimate the extent to which such optimal dosing strategies could reduce risk of treatment drop-out and relapse.

• ​​Racial Politics and Sentiment on Twitter: The Dynamics of Online Emotional Language 2021

  Colin Wayne Leach, Psychology, Africana Studies
  Courtney Cogburn, Social Work
  Sining Chen, Industrial Engineering and Operations Research
  Kathleen McKeown, Computer Science
  Susan McGregor, Data Science Institute

   

  Social media is a powerful means of individual expression, and collective consolidation, of people’s sentiment about the most important issues in our society.​ ​​This transdisciplinary​ ​project marries the latest advances in computational and statistical techniques of language use over time with social behavioral theories of emotion and stress to examine the temporal dynamics of ​tweets surrounding police killings of Black people and subsequent protest​s​ (e.g., Black Lives Matter).

• AI-Assisted Identification of Child Abuse and Neglect in Hospital Settings with Implications for Bias Reduction and Future Interventions 2020

  Aviv Landau, Data Science Institute;

  Desmond Patton, Social Work;

  Maxim Topaz, Nursing

   

  This team is developing an innovative artificial intelligence system to detect and assess risk for child abuse and neglect within hospital settings that would prioritize the prevention and reduction of bias against Black and Latinx communities.

• Data Driven Discovery of Latent Structure in Human Information Demand 2020

  Jacqueline Gottlieb, Neuroscience

  Vince Dorie, Associate Research Scientist, Data Science Institute

   

  In this project, online behavioral data will be collected from a large sample of participants, using a battery of tasks that probe different theories of how information is prioritized and used. This combined data set will allow an analysis of the latent factors that shape human-information demand while also unifying those theories.

• Detecting and Attributing Spatiotemporal Variations in Sources of Ground-Level 2020

  Ruth DeFries, Ecology, Evolution and Environmental Biology

  Arlene Fiore, Earth and Environmental Sciences;

  Jeff Goldsmith, Biostatistics

  Marianthi-Anna Kioumourtzoglou, Environmental Health Sciences

  Daniel Westervelt, Lamont-Doherty Earth Observatory

  John Wright, Electrical Engineering

   

  This team will develop methods to extract patterns from multiple datasets and identify the dominant sources of air pollution across India and how they vary in space and time. Their work is a step towards the overarching goal of informing effective clean air solutions and reducing public health burdens associated with exposure to air pollution in India.

• Gender and Racial/Ethnic Inequality in High-Skilled Labor Market: Gaining New Insights from Online Resume and Reviews Database 2020

  Kriste Krstovski, Data Science Institute

  Yao Lu, Sociology

   

  This team combines new sources of labor market data with data science methods to identify factors and environments that shape gender and racial inequality in high-skilled labor market. The team will chart long-term career trajectories of a large number of high-skilled American workers and examine gender and racial variations; and construct measures of company environment, especially that pertains to gender and racial equity, and assess its consequences for the career path of different groups of skilled workers.

• Interpretable Microbiome Dynamics in Liver Transplant Recipients 2020

  Itsik Pe’er, Computer Science

  Anne-Catrin Uhlemann, Medicine

   

  This team is developing methods for temporal analysis of gut microbiome compositions to better define the risk of infections in liver transplant recipients. They will integrate existing coarse resolution data with newly collected deep metagenomics and metabolomics data.

• Probabilistic Modeling of Intercellular Interactions that Drive Ferroptosis 2020

  Elham Azizi, Biomedical Engineering

  Jellert Gaublomme, Biological Sciences

  Brent Stockwell, Biological Sciences

   

  This team will develop probabilistic models to elucidate the role of intercellular interactions in driving susceptibility of treatment-resistant mesenchymal tumor cells to a newly discovered ferroptotic vulnerability, which could offer a therapeutic avenue to prevent survival of these cancer cells that are prone to metastasis.

• Molecular Mechanism of Treatment-Resistant Depression 2019

  Rene Hen, Neuroscience and Psychiatry
  Sergey Kalachikov, Chemical Engineering

   

  Major depressive disorder is a debilitating illness that affects more than 350 million people around the world. The most common treatments are drugs such as Prozac. About half of the patients who take the pills, however, do not respond to treatment. This team is thus trying to understand the molecular mechanisms of such treatment resistance. Ultimately, they would like to be able to predict which people will respond to antidepressant drugs before they begin treatment, and to develop new treatments that can circumvent antidepressant resistance in the millions of people who do not respond now to antidepressants.

• Data-Driven Modeling and Estimation of Li-Ion Battery Properties 2019

  Matthias Preindl, Electrical Engineering

  Alan West, Chemical Engineering

   

  This engineering team is developing a machine-learning model that can estimate a Li-Ion battery’s charge level with greater accuracy, aiming for an error rate of just one percent.

• Machine Learning to the Rescue: Eliminating the Dama Caused by the Tsetse Fly in Sub-Saharan Africa 2019

  Szabolcs Marka, Physics

  Zsuzsanna Marka, Physics

  Zelda Moran, Public Health;

  John Wright, Electrical Engineering

   

  This team is pioneering a machine-learning based imaging and sorting solution that aims to drastically reduce Africa’s tsetse population. The solution, which allows for the sorting of male and female tsetse flies, to support the Sterile Insect Technique, which the IAEA has used to eradicate tsetse populations in Zanzibar and other countries.

• Nudging New York: Using Data Science to Increase Healthcare Access in Underserved Communities 2019

  Marianthi-Anna Kioumourtzoglou, Environmental Health Sciences

  John Paisley, Electrical Engineering

  Kai Ruggeri, Health Policy and Management

   

  This research team intends to reduce missed appointments at community clinics by using big data and Bayesian machine learning techniques to understand why patients miss appointments and what can be done to help them keep them.

• Tracking Air Pollutants and Reconstructing 3D Scalar Fields from 2D Satellite Images via Machine Learning 2019

  Pierre Gentine, Earth and Environmental Engineering

  Marco Giometto, Civil Engineering and Engineering Mechanics

  Mostaf Momen, Civil Engineering and Engineering Mechanics

  Carl Vondrick, Computer Science

   

  This team is developing machine-learning models and improved satellite-imaging techniques that will help environmental officials locate and characterize hazardous pollutants in the lower atmosphere, allowing them to design strategies to mitigate pollution.

• A Game-Theoretical Framework for Modeling Strategic Interactions Between Autonomous and Human-Driven Vehicles 2018

  Xi Chen, Computer Science

  Sharon Di, Civil Engineering and Engineering Mechanics

  Qiang Du, Applied Physics and Applied Mathematics

  Eric Talley, Law

   

  This team is developing a fundamental framework using the game theoretic approach to model the strategic interactions of conventional human-driven vehicles and autonomous and/or connected vehicles. Other than technical advances, this project will also address the Trolley Problem (i.e., ethical sense development) in AV algorithm design.

• Enabling Small: Data Medical Research with Private Transferable Knowledge 2018

  Roxana Geambasu, Computer Science

  Daniel Hsu, Computer Science

  Nicholas Tatonetti, Biomedical Informatics

   

  This team is building an infrastructure system for sharing privacy-preserving machine learning models of large-scale, dynamic, clinical datasets. The system will enable medical researchers in small clinics or pharmaceutical companies to incorporate multitask feature models learned from big clinical datasets to bootstrap their own machine learning models on top of their (potentially much smaller) clinical datasets. The multitask feature models protect the privacy of individual records in the large datasets through a rigorous method called differential privacy.

• p(true): Distilling Truth by Community Rating of Claims on the Web 2018

  Trenton Jerde, Zuckerman Institute

  Nikolaus Kriegeskorte, Zuckerman Institute

  Nima Mesgarani, Electrical Engineering

  Chris Wiggins, Applied Physics and Applied Mathematics

   

  This team will build a complementary mechanism for web-based sharing of reasoned judgments to perform probabilistic inference on contentious claims with machine learning algorithms and bring rationality to the social web.

• Planetary Linguistics 2018

  Michael Collins, Computer Science

  David Kipping, Astronomy

  This team will build predictive models capable of intelligently optimizing telescope resources, and uncover the rules and regularities in planetary systems, specifically through the application of grammar induction methods used in computational linguistics.

• Predicting Personalized Cancer Therapies Using Deep Probabilistic Modeling 2018

  David Blei, Statistics

  Anna Lasorella, Pediatrics

  Raul Rabadan, Systems Biology

  Wesley Tansey, Systems Biology

   

  This team aims to model, predict, and target therapeutic sensitivity and resistance of cancer. They will integrate Bayesian modeling with recently developed variational inference and deep learning methods and apply them to large scale genomic and drug sensitivity data across many cancer types.