Text Mining Collaboratory

As attention to artificial intelligence (AI) mushrooms, a growing chorus across the private, government, and academic sectors is asking the humanities to help transform AI into a deeper and more socially responsible field. As humanities scholars and technologists in the humanities, we know that knowledge is never a finished output but rather something entangled with a continuing unfolding of meaning in the context of an ever-accumulating understanding of society and culture. By working together to create and examine new methods, code libraries, and theories for text mining, our lab hopes to reduce the societal and intellectual risks of contemporary text mining while bringing together humanists and scientists to develop novel, interesting, intelligent and responsible forms of computational knowledge. The research, development, and adoption of a broadly intelligent text mining language AI stands to benefit from a humanities-based approach to computing text-based knowledge.

The aim of the Duke Text Mining Collaboratory (TML) is to provide to the greater Duke community an opportunity for interdisciplinary collaborative work in text mining. Students, staff and faculty work together to research, develop, and translate language-based artificial intelligence within an arts and humanities framework. Founded in December 2017 by Patrick Herron, the purpose of the lab is to increase the adoption of text mining in the humanities while also creating new forms of text mining. Faculty, staff and students, graduate and undergraduate alike, learn, experiment and work together on specific research projects designed to meet the lab’s aims.