Master’s in Applied Data Science Winter 2025 Capstone Winners
The Winter 2025 Capstone Showcase for the Master’s in Applied Data Science (MS-ADS) program at the University of Chicago highlighted the exceptional talent and innovative thinking of its students. Fifteen teams presented their projects, addressing real-world challenges across various industries. The event culminated in the recognition of the Best in Show winners, whose projects exemplified the program’s commitment to practical, impactful data science solutions.
Learn more about the projects that won Best in Show below.
BEST IN SHOW WINNERS
Leap of Faith | Automatic Generation of ‘Quality Scores’ from Physician Notes
Presenters: Qing Chen, Heidi Abrahamson, Kevin Sianto
Faculty Advisor: Dmitri Sidorov
Project Topic: Manually reviewing physician notes for quality assessment is a costly, time-consuming, and error-prone process. The team leveraged Large Language Models (LLMs) to automate quality score generation, ensuring a faster, cost-effective, and accurate approach. This AI-driven solution reduces administrative burden, enhances efficiency, and allows healthcare professionals to focus more on patient care. By streamlining quality evaluation, the project demonstrates the transformative potential of AI in healthcare.
UChicago Data for the Common Good | AI for Health – Efficient Identification of Neighborhood Disorder
Presenters: Utsav Thota, Apoorv Anand, Christopher Marasco, Caleb Dimenstein
Faculty Advisor: Gizem Agar
Project Topic: This capstone team partnered with Data for the Common Good, a UChicago initiative that maximizes data’s potential to improve human health and reduce health disparities. This team developed an AI-driven solution to automate neighborhood disorder identification, replacing the manual process. This approach could cut man-hours by 98% and expand data collection 60X, making community health research faster and more scalable.
Fintech Entity | Unlocking Customer Insights to Optimize Lease-to-Own Conversions
Presenters: Hailey Kleban, Jack Feng, Simran Karamchandani, Juan Bautista
Faculty Advisor: Fan Yang
Project Topic: This team worked to predict which approved customers are most likely to sign a lease in a lease-to-own program. Using advanced machine learning, they analyzed customer behavior and financial patterns to help optimize outreach strategies and improve conversions.
Judges’ Insights
Mansi Rana, Senior AI Research Scientist at Uniphore and one of the judges, remarked: “Judging this capstone showcase was genuinely inspiring—the students displayed thoughtful innovation and an impressive depth of research, even after some of their milestones hit roadblocks. What stood out most was their ability to tackle real-world challenges with fresh perspectives and practical solutions, keeping production metrics like latency and cost in mind as well. The projects chosen felt both innovative and deeply relevant. I am confident that each and every presenter is a passionate learner, prepared to make a meaningful impact in their fields.”
Student Reflections
Reflecting on the capstone experience, 2025 MS-ADS graduate Simran Karamchandani, a member of one of the Best in Show-winning teams, shared: “Working on this capstone was an incredibly rewarding experience! Our team had the opportunity to thoroughly apply what we learned in class while also gaining valuable guidance from our client, which helped shape our approach. It was great to collaborate and refine our work with real-world input while bringing our own analytical skills to the table. The experience was extremely fulfilling, as we got a chance to see our efforts translate into a meaningful output that provided quality insights for our client.”
The Winter 2025 Capstone Showcase not only underscored the students’ technical prowess but also their ability to apply data science methodologies to solve complex, real-world problems. The MS-ADS program continues to foster an environment where emerging data scientists can collaborate, innovate, and make significant contributions across various industries.