Tuesday, October 18, 2016

After a year of attending the Computational Text Analysis Working Group at the UC–Berkeley D-Lab, which serves graduate students working in data intensive social science, our informally “embedded” librarian has created new opportunities for the library support of a burgeoning research community. Hennesy defines and gives examples of computational text analysis (also known as text mining), and illustrates why librarians are particularly well equipped to help with these new modes of research. He shares a few highlights: finding and accessing congressional records for a large-scale text analysis project, reviewing vendor text mining products, connecting with faculty to identify new corpora acquisitions, collaborating on text analysis documentation projects, and defining the legal boundaries around scraping various online text sources.