Nationally there is a disproportionate number of children of color (African American & Latino) in the child welfare system. Durham County is no different. However, reviewing this problem through the lens of data has not been done to formulate or implement possible solutions. Durham County Department of Social Services Child & Family Services would like to evaluate systems to identify where and how disproportionality and disparity are occurring. It is occurring at the entry point of Reporting child abuse and neglect? Is it occurring at the case decision? Is our reunification time different for African American children? Or Does it take longer for a child of color to achieve permanence through adoption? Organizing the data to show us our “hot spots” would facilitate further discussion and focus on solutions to an age-old systemic problem.

Faculty Lead: Greg Herschlag

Project Lead: Jovetta L Whitfield

A team of students will use visual reporting tools, such as Tableau or Power Bi, to create a dynamic dashboard that will enable investment professionals at Duke University Management Company (DUMAC) to review and better understand fund managers’ exposures and positioning across various dimensions.  Students will collaborate with teams at DUMAC to develop an intuitive, visual dashboard to help the investment team review individual and portfolio-level exposures across numerous asset classes.  This dashboard will be connected to DUMAC’s data warehouse, which refreshes on a daily basis as new data comes in for DUMAC’s public market accounts.

Finding similarity and node correspondence among networks has been widely applied in social network analysis, network de-anonymization, protein-protein interaction (PPI) network alignment, and shape matching in computer vision. A team of students led by Professor Jiaming Xu and Ph.D. student Sophie Yu will implement our recently developed graph matching and similarity scoring algorithms, test their empirical performance in real datasets, and further improve the algorithm design by incorporating domain knowledge. Multiple datasets will be explored such as Facebook ego network, Covid PPI network, Isobase PPI network, computer vision datasets, and Wikipedia article networks. A visualization app will be built to facilitate researchers and practitioners to apply graph matching in relevant applications.

Along with the Energy Data Analytics Lab and the Sustainable Energy Transitions Initiative, students will explore how machine learning can be developed to better identify and characterize energy infrastructure and scale up its application across geographies. Using synthetic data generation such as style transfer and/or 3D models to create images of artificial cities and communities, we will examine how well an algorithm trained on one set of training data is able to adapt to new places and types of energy infrastructure. This research will increase the speed and scale of assessment towards an automated, global assessment of energy infrastructure. The resulting data would empower decisionmakers and energy system planners to make better decisions for planning electricity system development for communities with limited electricity access or reliability.

The goal of this project is to use evaluation data collected on baseball athletes to make predictions about their on-field performance in competitive games. Evaluation data includes visual skills assessed with eye tracking, movement screening data and simulated batting performance metrics collected in virtual reality that are measured as part of the USA Baseball Prospect Development Pipeline. Using regression and Bayesian statistical analyses we will identify characteristics of these data that correlate with batting performance in order to inform scouts about the likely production of developmental prospects. The final product will be an application that uses an athlete's assessment results to produce performance summary graphs for the individual compared to other athletes and inferential models for the relationships between assessments and performance.

Project Leads: Marc Richard, Suhail Mithani, Greg Appelbaum

The Root Causes Fresh Produce Program, led by an interdisciplinary team of graduate and undergraduate students at Duke and UNC, leverages the power of student volunteers and community partnerships to address food insecurity in our community by delivering fresh and locally sourced produce to the doorsteps of patients in the Duke, Lincoln, and Samaritan Community health systems. Our program leadership is looking for team members to join us in exploring the use of data visualization and analysis to improve upon the processes that allow us to carry out our student-run service. Specific projects will include improving the program’s delivery route optimization software, improving the crop assessment and inventory management tools of small farmers and wholesale markets, as well as developing a dashboard integrating client location data and GIS information to identify location-specific service providers. Together, in collaborating with our team of interdisciplinary students as well as local Durham agencies, we hope to give you an opportunity to make an impact on our local community and food system while gaining a deeper appreciation for the role of data visualization and analysis through a population health intervention.

Project Lead: Willis Wong

A team of students led by members of the Duke IT Security Office will develop machine learning features that can be used to identify previously unknown web attacks.  Attacks targetting websites and web applications are continuous across the internet and protecting Duke's web infrastructure is part of the IT Security Office's mission.  The team will have a large dataset of logs from web traffic to explore possible features and evaluate their usefulness for identifiying new types of web attacks.  The results from this project may be incorporated into Duke's IT security infrastructure to help protect Duke's network in the future.

Project Lead: Eric Hope

The goal of this Data+ project is to apply and extend custom analytics solutions to discover how life remains resilient in extreme environments. An explosion of data has resulted from recent discoveries of tiny single-celled life hiding out in the most extreme places on Earth. These single celled creatures, or microbes, thrive in hot springs at boiling temperatures, in brine lakes saturated with salt, and in deserts once thought to be sterile. Studies in the Schmid lab at Duke have generated large data sets tracking how microbes grow and change their gene expression when faced with extreme stress. However, the majority of these data remain to be analyzed. In this Data+ project, students will apply and extend the statistical models to ask how extreme microbes grow and change during exposure to nutrient deprivation, extreme stress, and genetic mutations.

(photo credit: Peter Tonner et al., 2020 PLoS Computational Biology https://journals.plos.org/ploscompbiol/article/comments?id=10.1371/journal.pcbi.1008366 )

Project Lead: Amy Schmidt

Project Manager: David Buch

Who should get to decide what a utopian society looks like? After London was razed to the ground in the Great Fire of 1666, its reconstruction into the “emerald gem of Europe” was heavily influenced by the monarchy and aristocratic elites. In building a utopian epicenter focused on political and economic interests, immense sacrifices had to be made by London’s most marginalized citizens. A team of students led by Nicholas Smolenski (PhD Candidate, Musicology) and Dr. Astrid Giugni (Lecturing Fellow, English) will thus explore how London was rebuilt into a utopia; by employing topic modeling and applying the resulting lexicon to seventeenth-century architectural sketches, students will demonstrate how a language of progress became inextricably linked to its own image while also exposing the paradoxes entrenched in utopic representation. This project will additionally show how its framework can apply to current political discourse, as the tearing down of statues and monuments over the past three years has highlighted inescapable tensions between a governing power, a nation’s history, and its people.

Mental illness is often over-represented among the incarcerated population.  Durham county leaders are aware of this, and have taken many steps in recent years to support incarcerated people with mental illness and also to reduce incarceration in this population.  This work requires a community-wide effort, and Duke University Health System, as the major health care provider in the county, has a potential role to play. This Data + team will use data from the Durham County Detention Facility and Duke Health System to examine patterns of health-service utilization in the incarcerated population, including those with and without mental illness diagnoses.  We will use statistical methods such as longitudinal modeling to analyze the effects of the many interventions recently implemented, including increased mental health services, medication-assisted treatment for addiction, among others.  

Project Lead: Nicole Schramm-Sapyta

Project Manager: Ruth Wygle

Is there a right type and amount of consumption? The idea of ethical consumption has gained prominence in recent discourse, both in terms of what we purchase (from fair trade coffee to carbon off-sets) and how much we consume (from rechargeable batteries to energy efficient homes). Concern with the morality of consumption is not new to capitalist societies; we can find debates on the ethics of consumption in the Middle Ages and the Renaissance. A team of students led by Dr. Astrid Giugni (English, Duke) and Dr. Jessica Hines (English, Birmingham Southern College) will analyze approximately 60,000 Medieval and Renaissance digitized texts by performing topic modelling, which will allow them to track associations between the languages of consumer culture and ethical practice.The team will organize their results in a series of visualizations that trace the development of the ethics of ‘consumerism’ in terms of frequency of use, timeline of relevant events, authorship, location of publication and name of printer/printing press, and genres.