Topic Modeling Against a Corpora

Register through Hesburgh Libraries

Offered by the Navari Family Center for Ditital Scholarship.

Divide and conquer a corpus of texts in order to better understand it as a whole.

Topic modeling is a process of dividing & conquering a collection of texts in order to better understand the collection as a whole. Given a corpora of documents (books, articles, Web pages, etc.), topic modeling divides the corpora into sub-corpora, and each sub-corpora will be identified with a theme. This process is sometimes useful for identifying genres, authors, and/or subjects in a body of literature.

This hands-on workshop will demonstrate and facilitate the use of a free Java-based program called Topic Modeling Tool. Participants are expected to bring their own computer, and the computer is expected to have Java already installed, which it probably does.

Open to Graduate Students, Undergraduates, Faculty, Staff, Postdocs