Research

This section will briefly discuss some of the projects I’m currently working on at school. Currently the descriptions are brief, but I do plan to elaborate more in the future. However, if you’d like more detail on a specific project please email me(email link on side bar).

Transcript of the Future (ToF)

Collaborators: Ben Koester1, Sophia Cotignola2, and Tim McKay1
Physics, University of Michigan-Ann Arbor1
Economics, University of Michigan-Ann Arbor2

In this project we seek to understand how a student’s college experience can be better quantified and represented. Experience is a broad term and can have a lot of different interpretations. For this work experience is referring specifically to the courses taken by a student and interactions with other students via these courses. As the project continues this framework for experience may be expanded to be more general as access to more data sets become available. To explore questions around experience we use enrollment data at UM obtained via the Learning Analytics Data Architecture (LARC). We represent students and courses as a bipartite graph, which we call a co-enrollment network. Two students are connected if they take the same course and two courses are connected if a student takes both courses. We use this network to explore various questions around the structure of enrollment (e.g., does community detection find majors?). One of the corner stones of a liberal arts education is exposure to diverse ideas, we can use this network to measure a students diversity (we explore various definitions of diversity).

Understanding Investor Evaluation Models

Collaborators: Tom Kehler1, Scott Page2, Rada Mihalcea 3, and Kevin Quinn4
CrowdSmart, San Francisco, CA1
Center for the Study of Complex Systems, University of Michigan-Ann Arbor2
Electrical Engineering and Computer Science, University of Michigan-Ann Arbor3
Political Science, University of Michigan-Ann Arbor4

For this project we are working with CrowdSmart, an online collaborative investment platform, to understand various models used for evaluation of new ideas. Given numerical scores of a company, written explanations for the reasoning behind the scores, and amount invested, we’d like to know if we can statistically distinguish complexity from uncertainty. We explore the variance in numeric scores to the distances (and/or clustering) between corpus of reasons given. Also using this data we’d like to know if we can distinguish group think from wisdom of the crowds.

Measuring Similarity

Collaborators: James Sappenfield1, Karim Lakhani1, Scott Page2
Harvard Business School1
Center for the Study of Complex Systems, University of Michigan-Ann Arbor2

Description coming soon!

Social and Environmental Outcomes from Large-scale Land Acquisitions

Collaborators: Chuan Liao 1, Allison Kelly 1, Tim Williams2, Jonathan Sullivan1, Arun Agrawal1, Dan Brown1
School For Environment and Sustainability1
Industrial and Operations Engineering2

Description coming soon!