Bio

Yulin Yu is a PhD candidate at the University of Michigan School of Information. Her research interest is broadly in computational social science, data science, and innovation. Specifically, she studies the drivers of innovation across various domains, science, workplace, and art, by investigating creative strategies through the application and development of computational frameworks. Her work has been published in top-tier general science venues (e.g., PNAS) and computational social science venues (e.g., The Web Conference, ICWSM, CSCW), and she has received a special recognition award from IC2S2. Yulin’s research has also been featured in prestigious science communication outlets such as Science, ScienceDaily, and Inside Higher Ed. Through collaborations with industry partners like Microsoft, TAL, and Altmetric, her work has influenced industry products, including Microsoft Viva Topics. Previously, she interned with the Microsoft Research Office of Applied Research team.

Driving Innovation Through the Lens of Recombinational Novelty

Innovation often emerges from the recombination of existing knowledge, technologies, ideas, or any medium in a novel way. However, which novel recombinations are most fruitful and under what conditions remain unclear. Addressing this gap enables not only experts, ranging from scientists and artists to policymakers, to effectively identify and leverage key drivers of innovation, but also empowers the general public (e.g., science or product consumers) to more readily adopt useful innovations. Therefore, the primary goal of my research is to quantify and identify novel recombination strategies that promote the creation and adoption of valuable innovations.

With the rise of data-driven trends (e.g., using large-scale datasets to conduct user analysis) increasingly leading to disruptive innovations, my most recent work explores novel strategies for leveraging these datasets to advance scientific innovation (PNAS 2024, Working paper 2024). As we enter a data-driven world with abundant open-access data, empirical datasets are becoming increasingly essential for discovering new research patterns. My work addresses a crucial question: how can we efficiently and creatively use or reuse vast amounts of data to advance science, particularly in fostering impactful and disruptive scientific work?

My contributions to innovation-related research encompass the adoption of novelty in music through recommendation systems (ICWSM 2023), the reception of scientific novelty in public spaces (Working Paper 2024), and the social disparities in network diversity that impact workplace and scientific innovation (PNAS 2022, WWW 2023, WWW 2024). Collectively, my research has deepened the understanding of recombination novelty’s role in emerging areas of innovation and developed new computational methods to quantify novelty in science, workplace, and arts.