Loading…
TCDL 2022 has ended
If you run into any issues during the conference, please feel free to email us at info@tdl.org

Back To Schedule
Wednesday, May 25 • 10:00am - 10:50am
3A: An attempt at metadata enhancement through machine learning | If you want something done right, do it yourself: Do Graduate Assistants (GAs) help or hinder in Institutional Repository (IR) projects? | When You Say Nothing at All: UX and Digital Coll

Log in to save this to your schedule, view media, leave feedback and see who's attending!


Show & Fail (session type) | Short talks about programs, initiatives, workflows, or strategies you lead or participated in at your library that didn’t work out as hoped. This session type is new to TCDL and will celebrate risk-taking rather than success (approx 10 – 15 minutes per talk).

An attempt at metadata enhancement through machine learning
Speakers: Jason Long, Todd Peters

This presentation will share what learned about machine learning and applicability to generate metadata to enhance discoverability during a pilot project. Object detection through neural networks is a rapidly developing field. Using machine learning large sets of images can be analyzed, objects detected and classified. We used the pretrained models COCO, Inception, ResNet, VGG19, and Xception to classify objects in images in our San Marcos Daily Record newspaper negative collection. Our initial use of these models did not yield usable metadata, however it did provide a useful first step into machine learning and knowledge to develop future research.

If you want something done right, do it yourself: Do Graduate Assistants (GAs) help or hinder in Institutional Repository (IR) projects?
Speaker: Alexa Hight

In this show and fail, Alexa Hight will discuss the various projects for the Institutional Repository (IR) at TAMU-CC where either one or multiple Graduate Assistants (GAs) were enlisted. In the almost two years since the pandemic began, GAs have been tasked to help with IR projects during library closures, downtimes (such as summer), and a GA specific to help on Scholarly Communication projects was hired. In this time, I have somewhat learned how to manage a GA, I’m learning how to be a manager, I’m learning how to teach not only how to use DSpace, but also teaching the complexities and nuances of open access and what versions of works can be added to the IR, as well as what copyright or Creative Commons license we can apply. My experiences can help others new to managing an IR, as well as managing people (or trying to). I’ve also learned how to come up with metadata and other projects with short notice. I’m still learning how to balance supervision and quality control with my other responsibilities, as well as how to mentor and empower the GAs to be better employees and humans.


When You Say Nothing at All: UX and Digital Collections
Speakers: Shelley Barba, Kimberly Vardeman

Once a digital project has been “completed” and developers have moved on—perhaps even leaving the institution—how can the new caretakers decide how to best maintain the project? The TTU Digital Resources department originally developed Streaming Audio and Visual Experience (SAVE) as a digital collection to hold recordings of the School of Music’s graduate student recitals. In 2019, we realized that SAVE had been inaccessible from the institutional repository for at least a year. To correct this mistake and others made during the creation of this digital tool, the department reached out to the library’s User Experience department. We launched a collaborative project to gather user data and opinions to understand how Music faculty and graduate students used the recordings. This exploration would guide a decision about whether the library would attempt to improve SAVE or retire SAVE and create a new workflow for accessing recital recordings. Unexpectedly, our plan for interviewing users fell flat despite our best efforts to recruit participants from the collection’s target audience. This presentation is about how very off we were in our assumptions and how we are using this ‘fail’ to still move forward.

Moderators
avatar for Christina Kellum

Christina Kellum

Library Production Associate, University of North Texas

Speakers
avatar for Todd Peters

Todd Peters

Head, Digital and Web Services, Texas State University
avatar for Kimberly Vardeman

Kimberly Vardeman

User Experience Librarian, Texas Tech University
avatar for Alexa Hight

Alexa Hight

Scholarly Communication and Copyright Librarian, Texas A&M University-Corpus Christi
JL

Jason Long

Programmer Analyst I, Texas State University
avatar for Shelley  Barba

Shelley Barba

Digital Scholarship Librarian, Texas Tech University
Shelley Barba is the Digital Scholarship Librarian and Electronic Theses and Dissertation Curator at the Texas Tech University Libraries. Ms. Barba’s research has focused on digital libraries, metadata, and management. Her articles have appeared in several library journals, and... Read More →



Wednesday May 25, 2022 10:00am - 10:50am CDT