Skip to main content

RESEARCH

Creating dynamic visualizations to investigate research questions and to foster student learning outcomes.

DSC / Research

Below are select projects from the DSC's research and work. If you are looking for a particular project's information and don't see it, please feel free to contact us via the Contact Page and we will be in touch!

 

Tracking Disciplinary Trends: A Study in Anesthesiology

Department: Biomedical Informatics 
Researcher: Danny Wu, PhD
Dataset and Source: 22,267 academic abstracts between 2000 and 2013 from the American Society of Anesthesiologist Annual Meeting Archive 
Status: Completed
Publications: In preparation for publication at the Jornal of the American Medical Informatics Association (JAMIA)
DSC Team Members: Lindsay Nickels, Erin McCabe, Ezra Edgerton, Anubhav Maity, Sally Luken
Goals and/or Outcomes: Methods included unsupervised topic modeling, latent Dirichlet allocation (LDA), Multi-level Model of Models (MLMoM), the DSC data platform. Identified MLMoM as an efficient and useful machine-learning technique for topic modeling of unstructured data. Established need for a more robust machine-learning approach to large, unstructured datasets. Produced a standard method for examining and organizing research themes, trends, and patterns in an academic discipline.

Publications & Presentations