Overview

How can we teach a computer to recognize the genre or authorship of a text? What about the entities (people, places, things) within the text? Find out in this introductory workshop on popular classification techniques used in machine learning. Along with discussing some of the many use cases for classification—from spam detection to named entity recognition—we will also look at the mechanics behind frequently used classification algorithms. Finally, borrowing from a classic example of early statistical text analysis, we will build our own classifier to study the (formerly) disputed authorship of The Federalist Papers.

This workshop is designed for participants who have taken the “First Steps with Python” workshop or who otherwise have a general understanding of Python’s syntax and data types. If you missed or want to review what was covered in “First Steps with Python,” you can find the tutorial on the DHLab’s GitHub repository.

Instructors: Catherine DeRose (DHLab) and Doug Duhaime (DHLab)

Registration & Requirements

This workshop is open to all Yale students, faculty, and staff, but space is limited. To register, visit the YUL Instruction Calendar. If you have registered, you will be sent a Zoom link the day before the workshop. If you don’t receive the email or lose the link, please contact the Digital Humanities Lab.

Participants are asked to come to the workshop with Anaconda Python (version 3.7 or higher) already installed. If you have trouble with the installation, stop by the Digital Humanities Lab’s virtual Office Hours for help.

Be among the first to know when future workshops are announced by signing up for the DHLab’s newsletter.