Texts, Discourses and Data: the Humanities and Data Science

Description

This course examines computer-aided methods used in digital humanities and the social sciences for analysing discourses, an object of study that draws together multiple ways that language reflects and shapes social meanings. Within this context, it introduces concepts and methods for analysing natural language data and applies these through a series of practical lab classes. The first part of the course focuses on classic discourse analysis methods drawn from corpus linguistics, as well as the essential preprocessing steps used to prepare texts for a range of analytical purposes. In the second part of the course we study topic modeling, a technique for unsupervised, exploratory data analysis that has been widely used in digital humanities, and, finally, consider supervised text classification methods to identify discursive attributes such as sentiment, genre, or style.