ALCTS CaMMS Faceted Subject Access Interest Group Meeting
ALA Annual 2017
The Faceted Subject Access Interest Group (FSAIG) provides a forum to discuss the theory and use of vocabularies designed for faceted subject access. Join us in Chicago to hear three presentations involving the Faceted Application of Subject Terminology (FAST) headings.
Saturday, June 24, 2017
4:30 – 5:30 pm
Hilton Chicago, Continental B
OCLC FAST (Faceted Application of Subject Terminology) Project Update
Eric Childress, Consulting Community Manager, OCLC Research
The latest update from the OCLC FAST Project.
FAST and improving Resource Discoverability for Non-Roman Language Collections
Magda El-Sherbini, Head of Collection Description and Access Department, Ohio State University
Providing subject access to non-English materials using Library of Congress Subject Headings (LCSH) is a challenge not only for catalogers, but also for users who don’t know the structure of the LCSH. In many cases, the terms used in the LCSH are not equivalent in meaning to those in non-English languages. The current practice of OCLC to convert the LCSH to FAST terms opens new possibilities for improving resource discoverability and access to non-English collections. The presentation will address the following questions:
--Can MARC21 records be utilized to display non-English subject terms in the integrated library system (ILS)?
--To maximize discoverability of materials, can non-English terms be linked to FAST subject terms?
--Can users search the library catalog by subject terms in their preferred languages, e.g., Arabic, Chinese, Japanese, Spanish, or French?
FAST Times in Digital Repository Metadata Remediation
Rachel Jaffe, Metadata Librarian, University of California, Santa Cruz
Presently the Metadata Services department at UC Santa Cruz is engaged in a digital repository metadata remediation project in anticipation of both a possible DAMS migration and a transition to linked data. As part of our effort to control and reconcile the controlled, uncontrolled and out-of-control vocabulary terms appearing in our metadata, we are exploring the viability, feasibility and impacts of mapping our LC and other subject terms to FAST. Key issues that have arisen include:
--End user tasks: How do want to deal with our pre-coordinated LC headings? How will un-coordinating them impact search and retrieval? Do we want to replace LC terms with FAST, or provide both?
--Workflow: How do we train catalogers to assign FAST headings? Should catalogers supply FAST headings, or is it more expedient to use automated processes (e.g. OpenRefine, MarcEdit or OCLC’s tools) to convert assigned LC headings? How do we work around the limitations of particular tools?
--Community: What are other institutions doing? Has there been any assessment? Is faceted access just a trend or are we adding value? Is that value long-term value?
How to get FAST fast? Automating LCSH to FAST heading conversion with FAST Linked Data API
Lucas Mak, Metadata and Catalog Librarian, Michigan State University Libraries
Lisa Lorenzo, Metadata Librarian, Michigan State University Libraries
Within the Michigan State University Libraries’ Islandora digital repository, metadata development and maintenance is informed by best practices for discoverability and interface usability and design. Together, these factors led the MSUL digital repository team to implement FAST heading usage across all collections in the repository. FAST headings allow for succinct subject description and intuitive faceting as well as a cleaner display than complex, compound subject strings. In addition to use in display, FAST headings also lend themselves to conversion to linked data formats due to their more accurate assignment of URIs to individual subject strings.
In order to realize these benefits, an XSLT process was developed to convert LCSHs in metadata records into corresponding FAST headings utilizing OCLC’s FAST Linked Data API. An additional conversion was implemented specifically for the ProQuest’s UMI topical headings in our ETD collection. On top of the FAST subject string, URIs were also captured to enable our linked data experiment with schema.org and JSON-LD.