This workshop will explore using computational methods to analyze gender in movies. Movies lend themselves nicely to computation, since researchers can leverage scripts, visuals, or metadata (e.g. genre, year, cast members and production crew). For example, the Geena Davis Institute on Gender in Media used machine learning to automatically detect the gender of characters in top-grossing films, which allowed them to study gender metrics at scale, such as how many lines are given to different genders. We will spend the first part of the workshop covering studies like this one that use machine learning, natural language processing, and data science techniques.
After the overview, we will discuss the pros and cons of using computational methods for studying social phenomena. Then, we will provide attendees with a dataset of movie scripts (with a pre-prepared API for basic processing functions) and let them analyze it for potential patterns we brainstorm as a group. Those who do not know how to code can be paired with those who do. Given the range of content and the interdisciplinary material, we hope this workshop will appeal to anyone drawn to the topic, regardless of computational background.
Requirements: Attendees should bring their own laptop. Knowing basic Python and having Python 3 installed is a prerequisite for those who want to code at the end. For participants who don’t mind pairing up for the coding activity, no prerequisites are required.