Loading…
DLF Forum 2018 has ended
Welcome to the 2018 DLF Forum, Learn@DLF, and NDSA’s Digital Preservation 2018!

Learn more and register here.

#DLFforum #learnatdlf #digipres18 #DLFvillage 

OSF Repositories: DLF Forum + DigiPres18

Shared Notes Folders: Learn@DLF (bit.ly/LearnatDLF2018), DLF Forum (bit.ly/DLFforum2018) + DigiPres 18 (bit.ly/digipres18)
Monday, October 15 • 4:00pm - 4:55pm
#m4e: Topic Modeling and Machine Learning

Sign up or log in to save this to your schedule and see who's attending!

Presentations
 
A Shallow Dive Into Deep Learning
Bret Davidson, Kevin Beswick
NCSU Libraries, United States of America
 
Deep learning is becoming ubiquitous in our everyday lives. We’ll provide an overview of what deep learning is, dispel the hype around this technology, and talk about the opportunities we see for cultural heritage organizations in this space by sharing the work we’ve done thus far.
 
 
Future/Death of the Library: A Collaborative, Experimental, Introspective Dive into Digital Humanities
Rebekah Cummings, Anna Neatrour, Elizabeth Callaway
University of Utah J. Willard Marriott Library, United States of America
 
This presentation describes how two librarians and a digital humanities researcher explored the “future of the library” genre by assembling a corpus of literature and analyzing it through topic modeling. We will discuss how our project fulfilled research objectives, produced unexpected insights, and prepared us to support digital humanities research.
 
 
Topic modeling: human vs machine
Erin Wolfe
University of Kansas Libraries, United States of America
 
Reporting on work with The Black Book Interactive Project (BBIP) to look at automated text analysis processes, including topic modeling and data mining, on a corpus of African American literary texts, with a focus on race and race-related metadata elements as defined in a local schema.
 

Using Machine Learning & Text-mining for Scholarly Output: A Semi Automated Approach to Unlocking Knowledge and Value Latent in Unstructured Text
Darnelle Melvin
University of Nevada, Las Vegas, United States of America
 
This presentation introduces a research design aimed on identifying metadata themes published in research literature, and track how those themes have changed over time across knowledge domains. Utilizing machine learning, text mining, and data visualization tools, this study uncovers themes by performing textual analysis on article abstracts.

Moderators
Speakers
avatar for Rebekah Cummings

Rebekah Cummings

University of Utah Library
Marriott Library, University of Utah
BD

Bret Davidson

Associate Head, Digital Library Initiatives, North Carolina State University
avatar for Peter Leonard

Peter Leonard

Director, Digital Humanities Lab, Yale University
avatar for Darnelle Melvin

Darnelle Melvin

Special Collections and Archives Metadata Librarian, University of Nevada, Las Vegas
avatar for Erin Wolfe

Erin Wolfe

Metadata Librarian, University of Kansas Libraries


Monday October 15, 2018 4:00pm - 4:55pm
Messina 1+2+3

Attendees (125)