This project investigates how the structure of science has evolved over the past 125 years, with a focus on interdisciplinary research. We aim to understand whether the growing interdisciplinarity of scientific papers reflects an expansion in the number of disciplines, increasing differences between them, or both. The student will help map and quantify how disciplines have drifted apart over time.
Science is often imagined as a single, interconnected system of knowledge—but in reality, it’s more like a vast and growing universe. Over the last century, scientific papers have become increasingly interdisciplinary, combining ideas and methods across diverse fields. This trend raises a fundamental question: is interdisciplinarity increasing because disciplines are becoming more distinct, or simply because there are more of them?
In this project, we use large-scale publication data to analyze how the scientific landscape has changed over the past 125 years. We ask: How many disciplines can we identify at different points in time? How different are those disciplines from each other? And how do changes in the “distance” between disciplines shape the way researchers work together?
The student will work with publication metadata (journal titles, keywords, subject tags, citation patterns) and assist in building visual and quantitative maps of the scientific world over time. This will include learning to identify clusters of research, measure how close or far apart they are, and trace how those relationships evolve. The work is inspired by questions in sociology, network science, and the philosophy of science—and fits into a broader effort by the Connected Data Hub at UVA’s School of Data Science to understand how knowledge grows, fragments, and recombines.
This is a great opportunity for students curious about how science works behind the scenes, and how ideas move and change over time. No previous research experience is needed—we’ll provide training, mentorship, and a welcoming team.
Curiosity about science, data, or artificial intelligence
Proficiency with Python or R required
Ability to work collaboratively and ask questions
Interest in how different fields approach research
How to work with large-scale scientific publication data (OpenAlex)
Introduction to network science and clustering methods for mapping science
Skills in data wrangling and visualization
Scientific communication / presentation & collaboration skills
Advanced programing skills for data analysis (proficiency with Python or R required)