Supporting Name Authority Control in XML Metadata: A Practical Approach at the University of Tennessee

Abstract

While many different endeavors to support name authority control in Extensible Markup Language (XML) metadata have been explored, none have been accepted as a best practice. For this reason, libraries continue to experiment with the sche-ma, tool, or process that best suits their local authority control needs in XML. This paper discusses current endeavors to support name authority control in XML for digitized collections and demonstrates an innovative manual solution developed and implemented by the University of Tennessee Libraries to achieve this goal. Even though this method for authority control in XML metadata still relies on manual efforts, it effectively reduces time and research work by efficiently setting priorities, identifying critical descriptive areas in the digital transcriptions, and identifying the most appropriate biographical resources to consult. The effectiveness of this approach in improving the rest of the metadata production workflow is evaluated and presented. Soon after starting digitization projects, many libraries and other institutions often find that keeping track of name access points in the Extensible Markup Language (XML) is a huge challenge, regardless of the XML schema used. This is particularly the case in many types of digitized objects such as manuscripts, music, and other types of special collections where the number of personal names is exponentially more than the number of items digitized; the names are dispersed all over the digitized transcriptions; and information about these names is ambiguous, vague, and incomplete. However, no matter how difficult keeping track of name access points in digitized materials is, it is necessary in order to keep digitized objects retrievable. Access points not only help in the retrieval process of documents, but also help keep materials by the same creators or about the same subjects together. To keep a successful track of name access points in XML documents, libraries have been experimenting with many different endeavors to find an effective way to achieve this goal. So far the efforts created to support name authority control in XML metadata consist of (1) using XML schemas to encode authority data; (2) endeavors for shared, cooperative, national, and international XML name databases; (3) manual and automated conversion tools from Machine-Readable Cataloging (MARC) to XML; and (4) automated generation of authority control through especially designed systems. The problem with most of these endeavors is that they only address the issue of how to encode name access points utilizing XML authority schema; they do not address the issue of how to extract or harvest these names directly from the XML records and transform them into useful access points. The few endeavors that have tried, such as the systems for automated generation of authority control, have only been successful in extracting names from XML records but not in turning them into

Control de la información autoral y organizacional para la optimización del acceso y visibilidad científica

“…La cantidad de información que se recupera en las búsquedas realizadas en un entorno web, independientemente de su asertividad y relevancia, hace que el proceso de tratamiento informativo o desambiguación de la autoría sea cada vez menos práctico de realizar manualmente. Para Veve (2009), en cambio, la desambiguación automática de los nombres nunca podría ser realizada solo por soluciones tecnológicas, y que siempre sería necesario el ojo de un humano para completar el proceso con éxito. Para Elliott (2010), un enfoque híbrido, que trabaja con soluciones tecnológicas combinadas con la verificación humana, parece ser una excelente alternativa; sin embargo, es necesario que cada caso abordado tenga en cuenta sus particularidades.…”

Extending Name Authority Work beyond the Cataloging Department: A Case Study at the University of Nevada, Reno Libraries

“…Conversion is also not recommended: "The idea of converting from MARC authority records into records that use the local XML schema sounds appealing, but this method creates double work for the library." 14 With conversion, a library would still need to establish new names in the LCNAF to complete the process, and LCNAF records have the advantage of being shareable in a national database. "If many headings have to be locally established in XML schema following the rigorous LCNAF standards, then libraries may find establishing the headings directly in the LCNAF more worthwhile because other libraries can benefit from this authority work.…”

Microwork platforms as enablers to new ecosystems and business models: the challenge of managing difficult tasks

“…Matching names in large bibliographic databases is precisely non-trivial since names are often misspelt and the same person can be referred to in a variety of ways, notwithstanding the existence of homonymy. The difficulty lies in deciding whether different works with similar author names belong to the same person, or not: although automated methods are now often selected for matching and disambiguation due to limited resources (e.g., Smalheiser and Torvik, 2009;Cuxac et al, 2012;Gurney et al, 2012;Wang et al, 2012a), manual matching is considered to yield higher levels of accuracy (Veve, 2009). In our experiment, AMT workers were asked to replicate manual checks done in the process of building a database of Spanish author-inventors described in Maraut and Martínez (2014) 5 , who combined automated matching techniques with expert validation of dubious matches to identify more than 4,000 Spanish author-inventors.…”