Special Topics in Political Science (490-0-20)

Topic

Machine Learning in Political Science

Instructors

Gustavo Diaz

Meeting Info

Scott Hall 212: Thurs 2:00PM - 4:50PM

Overview of class

This elective surveys the field of machine learning from the perspective of the social and political sciences. The primary emphasis would be on algorithms used for prediction, be it in the context of supervised or unsupervised learning. The first half of the course will focus on core methods in the field, including linear models, tree-based methods, neural networks, and ensemble methods. The second half would focus on advanced developments, including techniques to process and analyze unstructured data, large language models, causal inference, and the role of AI in society. Students should expect a combination of theoretical readings from statistics and computer science, along with applications or extensions relevant to political science and the social sciences.

Registration Requirements

POLI_SCI 403: Linear Models or equivalent experience with regression models.

Students without the recommended background are welcome but should consult with the instructor before registering.

Learning Objectives

Understand, evaluate, and apply various machine learning techniques.

Develop the skills and background required to pursue future learning of machine learning methods.

Teaching Method

Seminar combines lecture, discussion, and live coding demonstrations.

Evaluation Method

• Biweekly explorations (semi-directed coding assignments)
• Project milestone reports
• Final project: Data and methods paper

Enrollment Requirements

Enrollment Requirements: Reserved for Graduate Students.