
Archive: 2026 Research Bazaar
The Data Science Research Bazaar brings together researchers from diverse fields who use data, computing, and software in their work. The 2026 theme—Meaning in the Metrics: Communicating Research Insights—highlights creative strategies for translating research into compelling formats, and it explores how research insights can be communicated to a variety of audiences, ranging from collaborators to policymakers to the general public. The theme highlights how well-designed metrics can make raw data easier to understand.
Good metrics help reveal patterns, show what matters, and guide interpretation. They enable researchers to share their findings through interactive dashboards, visual storytelling, audio and video production, and other media. Whether you’re building software, analyzing data, or communicating results, this event offers a space to connect, learn, and innovate.
The main 2026 Research Bazaar took place on March 19. Additional trainings and focused discussions were scheduled throughout March and April.
Main Bazaar Schedule
The main Research Bazaar was held March 19, 9am-4:30pm in the Discovery Building’s DeLuca Forum.
- 8am: Check-in (advance registration required)
- 9-10am: Welcome, introductory speakers
- Pallavi Tiwari, Associate Professor and Vilas Distinguished Achievement Professor, Department of Radiology, AI, Computational Imaging, and the Quest for Precision
- Ian Hutchins, Assistant Professor, The Information School, Rewiring Research Intentionally: The Metrics We Choose Shape the Science We Get
- 10am-12:30pm: Lightning talks (three rounds with discussion breaks – presenters below)
- 12:30-1:30pm: Lunch
- 1:30-3pm: Ideas into Action panel
- Jenny Bratburd, Researcher and Outreach Program Manager, Center for Sustainability and the Global Environment
- Edith-Marie Green, Population Health Sciences, Center for Demography and Ecology
- Anna Haensch, Associate Director, Digital Scholarship Hub, UW–Madison Libraries; Research Associate Professor, Data Science Institute
- Michael Notaro, Director and Senior Scientist of the Nelson Institute Center for Climatic Research
- Daniel Pimentel, Assistant Professor, Department of Biostatistics and Medical Informatics and Wisconsin Institute for Discovery
- Moderator: Karen Schloss, Associate Professor, Department of Psychology
- 3-4:30pm: Poster session (presenters below)
Lightning Talks
Anton Babkin
Tidy public data: A standardized interface for R
Dhananjay Bhaskar
Learning Morphodynamic Trajectories for Embryo Assessment in IVF
Jennifer Bird
Data into Gems: An AI workflow for accessible data communication
Sam Blackburn
Floody, muddy, and green: An R Shiny app to teach data literacy and freshwater science in the classroom
Cynthia Burnson
Bringing actionable data to child welfare policymakers through event analysis and process mining
Daniel Ruiz de Castilla
Navigating chemical space: Reactivity descriptors for method evaluation
Chris Endemann
Supporting ML/AI work at scale at UW–Madison
Edith-Marie Green
Dementia research impacts us all: Tools for communicating research to mixed audiences through a health sciences lens
Josiah Hanna
What do we mean by “reason” in large reasoning models?
Nicholas Keuler
How the Statistical Consulting Group can help you find meaning in your metrics
Emma Longo
Computational Communication: How Fossil Fuel Companies Frame Environmental Issues in Annual Sustainability Reports
Abe Megahed and Dev Ahluwalia
Encouraging empathy using chatbots
Steve Meyer
Applying machine learning computational techniques to library subject headings
Maria Oros
The UW–Madison Open Source Ecosystem: 2024 Campus Survey Results
Zekar Otles
Exploring Azure low-code solution for meteorological data analysis
David Skrovanek
Weather to watts: Using weather data to optimize the grid
Roddy Taing
Consequences of model misspecification of adjacency spectral embeddings
Ziyan Wu
Untangling complex contaminants with surface-enhanced Raman spectroscopy and deep learning
Jiaqi Yang
Seeing individual tree with one AI: UniTree for label-free tree crown mapping
Sijia Yang
Persuasive AI for public health messaging
Zhou Zhang
An integrated deep learning modeling framework for county-level crop yield prediction in support of USDA NASS operation
Wenxuan Zhao
FLIM Playground: An interactive, end-to-end graphical user interface for analyzing single cells with fluorescence lifetime imaging microscopy
Posters
Yury Bukhman
Annotation of cell types in single cell RNA-seq datasets data
Anna Chinni
Semantic distance predicts ability to interpret texture meaning in information visualizations
Matthew Christie
Vertebrate: Open backbones for data architecture
Marek Estrada
Measuring LLM ambivalence: A psychometric approach
Yijing Gong
Optimizing dairy cow replacement decisions using reinforcement learning in an evolving stochastic environment
Sarah Graves
UAS lidar enables quantification of forest coarse woody debris to support forest carbon management plans
Tim Gruenloh
Is ChatGPT-4o racially biased for medical diagnosis?
Martin Guerra
What metric to optimize for suppressing instability in a Vlasov-Poisson system?
David Hardy
Actinorhodopsin pentamers are phospholipid scramblases: A molecular dynamics approach
Han Hu
Optimal maintenance polices for assembly lines with stochastically deteriorating machines and finite buffers
Christian Jorgensen
Interpretable visualizations of data spaces for classification problems
Martin Kaehrle
Who can use restricted-access species data?
Christina Koch
CHTC: Your home for research computing
Kailan Kraft and Cole Delong
Modeling brain connectivity to determine consciousness states
Chloe Kusser
Machine learning prediction of soil organic carbon fractions to assess global soil carbon sequestration potential
Shixiao Liang
Real-time lane-level crash detection on freeways using sparse telematics data
Changwon Lee
Sparsest neural network discovery via mixed-integer linear optimization
Jordan Loeffler
Developing low-dimensional order parameter to indicate crystallization phenomena
Gabriel McMahon
Children’s allergy and asthma data repository (CADRE)
Ella Mitchell
Food equity access simulation technology
Steven Moen
I-ACRE: An autoregressive conditional extremes model for incomplete panel data with applications in exoplanet detection
Marine Nalbandyan
Real-world evidence to advance community health (REACH)
Olivia Nygaard
Using Farm2Facts to strengthen sustainable and equitable food systems through data collection, analysis, and science communication
Aurod Ounsinegad
Identification of early-stage risk indicators of ovarian cancer with explainable machine learning in electronic health records
Harsh Rathore
Machine learning analysis of fast-Ion losses in stellarator experiments
Ben Rush
Adjudication of incidental findings from 18,000 chest CTs by CoReadAI, an LLM, and human readers
Eric Schultz
Practical Pitfalls of Active Learning for Protein Fitness Optimization: A GB1 Benchmark
Jennifer Sciana
Responding to failure: Developing a playful assessment of student mindset
Renee Sutkay
Improving healthcare value through chronic condition management
Roddy Taing
Consequences of model misspecification of adjacency spectral embeddings
Zezhong Tian
Mitigating NDVI saturation in imagery of dense and healthy vegetation
Xiaoyu Wang
Knowledge-guided machine learning model with soil moisture for corn yield prediction under drought conditions
Mingda Wu
Why phenology matters: Unlocking stage-dependent remote sensing signals to improve corn yield prediction across the U.S. Corn Belt
Runxiong Wu
What makes a good low-dimensional latent space? Generation, classification, and dimension reduction revisited
Zifan Xu
CNN and transformers for spectral reconstruction: A metric-driven interpretation
Tong Yu
A multimodal generative adversarial network for near-daily alfalfa multispectral image reconstruction
Extended Bazaar Events
In 2026, Extended Bazaar events were held in March and April.
March 23
Presented by Olivia Nygaard, Kaufman Lab
Collaborators: Alfonso Morales, Planning and Landscape Architecture; Abe Megahed and Steve Wangen, Data Science Institute; and Garrett Smith, TRAD
This discussion will explore how data-driven tools can generate meaningful insights that support food access and equity decision-making. Using Farm2Facts and FEAST (Food Equity Access Simulation Technology) as case studies, the session will examine how metrics move from data collection to interpretation, modeling, and application. Farm2Facts collects standardized data from farmers’ markets, including metrics related to sales, food access programs, vendor participation, and community engagement. FEAST builds on similar data sources through computational modeling and simulation to evaluate food access scenarios and equity outcomes. Together, these tools highlight complementary approaches to measuring and interpreting food system data.
The session will feature brief presentations from collaborators in data science and TRAD, followed by facilitated discussion and interactive activities. Attendees will examine real examples of metrics, visualizations, and model outputs and discuss challenges such as data quality, interpretation, scalability, and usability across disciplines and audiences. Participants will leave with a clearer understanding of how interdisciplinary collaboration strengthens applied data tools and how metrics can be transformed into actionable insights for food systems research and practice.
Presented by Lisa Abler, Libraries
Collaborators: Emma Bekele, Savannah Carr, Lauren Scanlon, Heather Shimon, and Hannah Swan, Libraries
Make a zine with us! Zines (bite-sized, handmade magazines) are creative, tactile expressions of an idea, experience, or point of view, usually designed to be shared and build community. Zines offer researchers a unique, artistic medium to process a complicated idea or to translate research for new audiences. This hands-on workshop will cover different kinds of zines, how to communicate a topic through visual design, and resources for further exploration. We’ll have a warm-up drawing activity and then guide you to create a page that will be collated into a community zine. Each participant will receive a copy of the finished community zine after the workshop. Zine-making supplies will be provided.
Presented by Ben Rush, Radiology
Collaborator: Anne Lynn Gillian-Daniel, Materials Science and Engineering
This interactive workshop is designed for scientists, trainees, and professionals who want to sharpen their ability to communicate effectively and adaptively. In academia and industry alike, ideas and stories often need to be shared in real time with diverse audiences—from colleagues and collaborators to decision-makers and the public. This session provides a lively, low-stakes space to build confidence in handling the unexpected. Through applied improv and playful practice, participants will experiment with spontaneous speaking, responding to mistakes, and refining how they connect with others. Attendees are expected to participate.
Goals:
- Increase your comfort communicating science by practicing how to adjust and pivot when the unexpected happens
- Develop various versions of your research overviews and stories for different audiences and scenarios
- Foster new connections and build community with other participants.
This session uses applied improv principles and interactive communication games to expand participants’ science communication toolkits. Activities are designed to meet participants where they are—whether graduate students honing their presentation skills or mid-career professionals looking to refresh their communication style. The workshop emphasizes practice over performance: participants are encouraged to engage at their own pace, laugh at mistakes, and support one another in experimentation.
April 6
Presenters: Sarah Stevens, Data Science Hub
This workshop lesson is an introduction to making interactive data visualizations in Python. Learners will wrangle data into the proper format using pandas library, create visualizations using the Plotly Python library, and display these visualizations and create widgets using Streamlit. Foundational knowledge in Python, such as that covered in the Software Carpentry Python lesson, is needed to follow along with the training.
Presenters: Ryan Bemowski, Data Science Hub
This workshop is an introduction to the R Shiny interactive dashboard package. We will use R via RStudio to build an interactive dashboard and upgrade it throughout the workshop. We will focus training specifically on Shiny and associated interactive packages. Working knowledge of R and RStudio is expected. You may also join this session to observe without participating.
April 16
Presenter: Chris Endemann, DoIT Research Cyberinfrastructure
As AI systems become more embedded in research workflows, they are evaluated using an expanding set of metrics including accuracy, speed, cost, energy use, explainability, adoption, and more. These metrics often compete with one another, and priorities are shaped by practical constraints around resources, scale, and time-to-results. This session focuses on how researchers choose and balance metrics when using AI in applied settings, examining how these choices influence trust, deployment decisions, and the feasibility of sustaining AI workflows over time. Rather than proposing a single evaluation framework, the discussion aims to surface real constraints, tradeoffs, and open questions researchers face when deciding what “good enough” looks like in practice.
Presenter: Steve Moen, Statistical Consulting Group
Collaborators: Cécile Ane and Nicholas Keuler, Statistics
High-quality experimental designs are crucial in science. In this workshop, four common ones are discussed – completely randomized designs (CRD), complete randomized block designs (RCBD), RCBDs with subsampling, and split-plot designs. Specific examples that tie to real-world analyses across disciplines are emphasized.
In addition, the workshop will include a statistical consulting demo to showcase the statistical consulting process, where scientific questions are discussed and advice is provided at no additional cost. If you are interested in taking part in a live consulting meeting, please bring information related to your research to facilitate a productive discussion. Participation is optional. Depending on demand, not all interested attendees will get to participate.
Sponsors
Thank you to our campus and industry sponsors for the 2026 Data Science Research Bazaar: Data Science Institute, Division of Information Technology, Libraries, Nelson Institute for Environmental Studies, School of Computer, Data and Information Sciences (Department of Computer Sciences, Department of Statistics, iSchool), School of Medicine and Public Health, the Wisconsin Alumni Research Foundation, and the Wisconsin Department of Revenue.

Data Science Institute
Department of Computer Sciences
Department of Statistics
Division of Information Technology
iSchool
Nelson Institute for Environmental Studies
School of Computer, Data and Information Sciences
School of Medicine and Public Health
UW–Madison Libraries


Planning Committee
The Data Science Research Bazaar would not be possible without the efforts of Planning Committee members who represent a variety of different fields and disciplines using computing and data science. This committee meets every 1-2 weeks in the months leading up to the Bazaar to plan, promote, and execute the event.
The members of the 2026 Planning Committee are listed below. We appreciate the efforts of these individuals and thank their respective departments/employers for contributing their time.
- Trisha Adamus, SMPH – Ebling Library
- Ryan Bemowski, Data Science Hub
- Cris Carusi, Data Science Institute
- Michael Ferris, Computer Sciences
- Erwin Lares, Research Cyberinfrastructure
- Mary Murphy, DoIT + Research Cyberinfratructure
- Lisa Play, Libraries
- Heather Shimon, Science and Engineering Libraries
- Sarah Stevens, Data Science Hub
- Ariella Thompson, Botany
View Recorded Sessions From the 2026 Research Bazaar
Click the button below to browse a playlist of all available session recordings (workshops, discussions, and presentations).
Previous Research Bazaars
The Research Bazaar is an annually occurring event, hosted by the Data Science Hub, with the mission of connecting researchers and data science professionals around diverse topics. The websites for previous Research Bazaars, beginning in 2021, are linked below.
2025 Research Bazaar: AI and ML in Research: Navigating Opportunities and Boundaries
2024 Research Bazaar: Data in Action
2023 Research Bazaar: Information Insights: Shaping Futures With Data and Computing
2022 Research Bazaar: Data and Communities: Sharing and Protecting Data