Posters

Starts at
Mon, Oct 21, 2024, 17:30 EDT
Finishes at
Mon, Oct 21, 2024, 18:30 EDT

Presentations

OpenWEMI - A minimally constrained vocabulary for Work, Expression, Manifestation, and Item

Authors: Karen Coyle, Phil Barker, Sean Petiya, Ross Singer, Graeme Williams

The concepts of Work, Expression, Manifestation, and Item introduced in the FRBR document are specific to library catalog applications. Yet these have been employed in metadata for resources quite different from those in the library bibliographic catalog, often ignoring some of the restrictions built in to the original definition. OpenWEMI retains the basic concepts of Work, Expression, Manifestation, and Item, but defines these without reference to the library use case. OpenWEMI classes and properties can therefore form the basis of description of any created resources.
  • Karen Coyle

    DCMI

    Karen Coyle is a librarian specializing in metadata standards. She has served on standards efforts including the MARC standards group (MARBI), NISO committee AX for the OpenURL standard, W3C's SHACL specification, and was an ALA representative to the e-book standards development that led to the ePub standard. She is currently investigating the possibilities offered by the semantic web and linked data technology, working with the Dublin Core Metadata Initiative. Karen's publications are available on her web site, https://kcoyle.net.

Managing Metadata Mapping

Authors: Trevor Stratton

Metadata mapping plays a pivotal role in the accessibility, management, and discoverability of digital resources. As the digital landscape continues to expand, service providers face the significant challenge of maintaining metadata quality while maximizing the level of discoverability of digital collections. This poster presentation seeks to explore the nuanced dynamics of metadata mapping within the context of library systems and digital collections.
  • Trevor Stratton

    A.T. Still University and University of Missouri-Columbia

    I graduated with a BA in History from the University of Missouri-Columbia in 2022. I also graduated with an MS in Library and Information Science as well as an MA in History from the University of Illinois Urbana-Champaign in 2024.

    I served as the Metadata Services Graduate Assistant for the University of Illinois Library for two years. I loved this role and it is what led me to pursue a career as a metadata librarian.

    I am currently the Metadata Librarian at A.T. Still University and in my first semester as a PhD student at the University of Missouri-Columbia.

Linked Data Bibliographic Description Based on User Search Behavior

Authors: Sangoh Na, Inkyung Choi, Yeonhee Park, Jongwook Lee

This study examines how librarians in Korean academic libraries perceive the usefulness of bibliographic enhancement data (BIBED) elements in cataloging tasks. Through a survey, it evaluates their perceptions of these elements' utility in scenarios involving searches for known items, unknown items, and related items. Additionally, by comparing various data models/metadata schemas, it proposes potential application profiles for book descriptions. This research highlights the importance of integrating user-centered elements into library catalogs and underscores the role of linked data technology in better meeting user needs.
  • Sangoh Na

    Kyungpook National University

    Sang-oh Na is currently the Integrated Ph.D. student in the Department of Library and Information Science at Kyungpook National University. His research interests encompass various aspects of library cataloging, with a particular focus on metadata and linked data.

Beyond OpenRefine: Assessing and Improving Name Reconciliation Strategies for Linked Data

Linked data is classically characterized as a transition from "strings" to "things," referencing the need to transform string literals into unique resource identifiers (URIs). This transformation, generally referred to as reconciliation, is one of the core activities required to create linked data. However, reconciliation can be fraught with issues such as false positive matches that require significant manual intervention. Therefore, improving reconciliation, by making it more efficient and reliably accurate, is necessary. This poster compares the performance of OpenRefine, a widely used tool for reconciliation, to that of a custom pipeline using Python and GPT-4o for name authority reconciliation. The preliminary results show that the multi-modal custom pipeline significantly outperforms OpenRefine in accuracy of automatic name matching, but additional engineering is required to improve processing performance.
  • Jack O'Malley

    Frick Collection

    Jack O'Malley is the Metadata Lead at the Frick Art Reference Library, where his primary responsibility is to describe and enrich access to digital resources. Jack is a member of the ARK Alliance Outreach Committee, a global community for promoting and sustaining open-source persistent identifiers, and has previously served on committees at the Metropolitan Library Council, the Society of American Archivists, and the Pratt School of Information. Prior to his employment at the Frick, Jack was a fellow at the New York Public Library and the Pratt Institute Archives.

AI and Metadata: Bridging the Gap between Digital Images and Human Understanding

Authors: Marina Morgan

As digital collections continue to grow, artificial intelligence (AI) tools have been used to generate titles, descriptions, and keywords for digital collections in libraries. This poster explores whether four AI models (Claude, Gemini, Copilot, and ChatGPT 4.0), can effectively provide descriptive titles, and whether librarians and archivists can bridge the gap between digital images and human understanding, enhancing metadata quality and usability. This poster also showcases each tool’s prospects for AI-driven metadata for title generation, highlighting the transformative potential of AI within digital collections.
  • Marina Morgan

    Florida Southern College

    Marina Morgan is Head of Technical Services and Senior Metadata Librarian at Roux Library, Florida Southern College. In her role, she supervises the Technical Services operations, develops procedures for processing all library materials, serves as the ILS and ProQuest ETD administrator, manages the Institutional Repository, and co-manages the Digital Collections.

    Before coming to Florida Southern College, Marina taught a wide range of metadata and cataloging courses in Canada at Seneca College - Library and Information Technician Program in Toronto, ON and at Mohawk College - Library and Information Technician Program in Hamilton, ON. As a librarian she worked in highly collaborative environments on various special collections and digital scholarship projects, while cataloging, promoting, and helping develop several digital collections.

Future-proofing Metadata at Los Alamos National Laboratory

Authors: Mark Offtermatt, Megan Rose Kilidjian, and Laura McGuiness

Los Alamos National Laboratory’s (LANL) National Security Research Center (NSRC) serves as the Laboratory’s classified library. It is home to a distinct collection of technical documentation spanning the Laboratory’s history. The NSRC supports the Laboratory mission, ensuring national security through best practice data and information delivery. The researcher’s need for immediate information access and storage has resulted in a lack of coherent metadata, generated by siloed working groups, across a range of disparate systems. The Laboratory’s Metadata Initiative (MI) consisting of library and information science professionals, assembled in 2023 to begin standardization across the complex. The initiative is engaged in several projects that will promote consistency across systems, driven by compliance with the National Archives and Records Administration (NARA) and the Dublin Core Metadata Initiative (DCMI). While compliance with best practices will enhance search consistency and interoperability throughout LANL and the broader enterprise, the unique nature of Laboratory materials requires fulfilling specific metadata needs as well. In this poster, we will discuss the goals and challenges of balancing researcher needs and varied access systems with interoperability and standards compliance.
  • Mark Offtermatt

    Los Alamos National Laboratory

    Mark Offtermatt Joined the Los Alamos National Laboratory's National Security Research Center in early 2022 as a senior librarian and digitizing archivist. With over a decade of experience in both public and academic service, he currently serves as the AV Digitizing Archivist, and Lead Technical Librarian for the NSRC's Metadata Initiative. He has a Bachelor of Arts in Russian and Eurasian Studies from The Evergreen State College, and a Master's degree in Library and Information Science from the University of Washington.
  • Megan Rose Kilidjian

    Los Alamos National Laboratory

    Megan Rose Kilidjian is a motion picture film archivist at the National Security Research Center, Los Alamos National Laboratory’s classified library and archive. Previously, she’s held positions as an audiovisual preservation project manager at George Blood Audio/Video/Film, and as the public media digitization archivist at New Mexico PBS. She completed her master’s degree in library and information science with a concentration in archives management at Simmons University. She currently serves on the Federal Agencies Digital Guidelines Initiative (FADGI) Motion Picture Film Working Group, and Los Alamos National Laboratory's Metadata Initiative Committee.

Optimizing metadata enhancement workflow of current research information system via CrossRef – pilot study for a real-life project

Authors: Kin Lok Rocky Mak, Richard Parish

This pilot study aims to explore the possibility of reducing human intervention in assuring metadata quality of a current research information system via CrossRef metadata ingestion. 5 key Dublin Core elements have been selected for review and a significant discrepancy in metadata accuracy between fields is observed. Further study is needed to confirm possibility of automation.
  • Richard Parish

    Anglia Ruskin University

    I have worked in the library field for 14 years, starting off in public library acquisitions before moving to academic content delivery and subscription management. Just over a year ago, I moved to the Research Services team at Anglia Ruskin University, where I help to run and maintain the institutional repository and advise researchers on open access. All of my roles have been heavily involved in metadata creation and curation, plus I have worked on a number of large-scale system migrations, so I am strong proponent of metadata standardisation, data integrity and quality control.

Compliance Rating Scheme: Data Provenance for Dataset Use in Generative AI Applications

Authors: Matyas Bohacek, Ignacio Vilanova

Generative Artificial Intelligence (GAI) has experienced exponential growth in recent years, partly facilitated by the abundance of open-source large-scale datasets. These datasets are often built using unrestricted and opaque data collection practices. While most literature focuses on the development and applications of GAI models, the ethical and legal considerations surrounding the creation of these datasets are often neglected. Specifically, the information about their origin, legitimacy, and safety often gets lost. To address this, we conceptualize the Compliance Rating Scheme (CRS) as a tool to evaluate a given dataset’s compliance with a set of practical principles, enabling developers and regulators to gauge and verify the transparency, accountability, and security of these resources. We open-source a Python library built around these principles, allowing the integration of this tool into existing pipelines.
  • Matyas Bohacek

    Stanford University

    Matyas Bohacek is a student researcher at Stanford University, advised by Professor Hany Farid from UC Berkeley. His research focuses on generative AI, deepfake detection, and other problems at the intersection of AI and media forensics.

A Community-Centric Approach to Bilingual Metadata and Ethical Representations in Oral History

Authors: José J. Morales Benítez1, Natalia Hernández Mejías, Jaquelina E. Alvarez, Ricia A. Chansky

This poster describes the process a team of librarians followed to create and implement a metadata design for a newly established Oral History Lab. The design prioritizes dignified representations of interview participants and supports their autonomy by including them in the description of their stories. The poster also covers the ethical considerations that have guided the work.
  • Natalia Hernández Mejías

    University of Puerto Rico at Mayagüez

    Natalia Hernandez has a Master's degrees in History and in Library and Information Sciences from the University of Puerto Rico, Río Piedras Campus. She has worked in a wide variety of librarianship roles in academic, public, and special libraries, from reference services to workshop instructor, as well as project and grant management. Her interests in the field include access to information, cataloging, metadata, and community oral histories. From 2020 to 2023, she worked as project manager for the Mellon Foundation’s digitization initiative at the General Archives of Puerto Rico. Recently, she joined the library team at the UPR in Mayaguez, as a metadata specialist, where she works organizing assets of the institution’s Oral History Lab.
  • José J. Morales Benítez

    Universidad de Puerto Rico, Recinto de Mayagüez

    José J. Morales Benítez is the Research Services Librarian in the Graduate Research and Innovation Center (GRIC) at the General Library of the University of Puerto Rico at Mayagüez (UPRM). His role is to provide support services to graduate students and faculty in all stages of the research process. As part of his duties at GRIC, José designs and delivers workshops on topics such as data management, data sharing, scholarly communications, open access publishing, and discovery and evaluation of information sources. He also creates online guides and videos and offers individual consultations on these topics. Additionally, José helps manage Scholar@UPRM, which is UPRM’s institutional repository, providing support to graduate students and faculty in the self-deposit process and checking for metadata consistency. Besides working at GRIC, José collaborates with UPRM's Oral History Lab (OHL), having led the initial design for the creation and management of the OHL's digital collections. This collaboration has included establishing metadata systems, establishing workflows for the intake and processing of oral history interviews, and working to maximize dissemination and access to the oral histories. José has also helped organize numerous workshops presented by the OHL addressing topics such as digital preservation, metadata creation, and the development of oral history collections and community archives.

Encoding Reparative Description: Promoting Archival Metadata Repair through Visualization and ArchivesSpace

Authors: Jesse A. Johnston, Max Eckard, Gideon Goodrich, Ella Li

Over the past few years, more archives and archivists have been working on enhanced description projects that can address past inequities, erasure, or incorrect representations in description. Stemming from the “ReConnect/ReCollect” project at University of Michigan, which has surveyed the extent and legacy of colonial collections extracted from the Philippines since the late nineteenth century, we report on work to analyze more than two hundred finding aids with the development of Python-based analysis tools. We demonstrate how automation can help to expand the project of reparative description. The poster reports on how archivists, faculty, and students, worked across the University to aggregate finding aid metadata and analyze that descriptive information. We present information about the code that we used, and the work to update and implement computational tools to work with the open-source archives information management application, ArchivesSpace, in order to aid the goals of reparative description workflows in archives.
  • Jesse A. Johnston

    University of Michigan School of Information

    Jesse Johnston is a Clinical Assistant Professor at the University of Michigan School of Information. Johnston has been active in the cultural heritage field as an administrator, librarian, archivist, and civil servant. As senior librarian for digital content at the Library of Congress, he managed policy development and training initiatives at the Library's Digital Collections Management division. Prior to that, he served as a senior program officer for preservation and access at the National Endowment for the Humanities.
  • Max Eckard

    University of Michigan. Bentley Historical Library

    As head of the Bentley's Curation team, I represent the team on the Bentley Leadership Team and lead the Bentley's holistic approach to curating archives of all genres and formats.