One Concept, One Term, Good Practice but How to Achieve? —Improving Facet Values Quality for Samuel Proctor Oral History Collection, Hosted by the University of Florida Digital Collections

Abstract Faceted search, also known as dynamic taxonomies, is a popular feature applied to digital collection sites. Appearing as clickable labels, facets facilitate search result refinement and content browsing. To achieve the utmost efficiency, faceted search requires each facet value to represent a single concept–that is, one controlled vocabulary term represents one concept. However, in reality, this status is hard to achieve. An example of this can be seen in the digital collection of Samuel Proctor Oral History Program, a high-profile international resource hosted by the University of Florida Digital Collections. The Topical Subject and Genre terms appear with many vocabulary control issues: the same concept is often expressed with different terms; the same term appears with different spelling variations; and/or outdated terms mingle with more up-to-date ones. Additionally, compound terms that represent multiple concepts prohibit the grouping of content that share individual concepts. In short, a great deal of improvement will be needed to optimize the faceted search. To address these issues, the Digital Support Services department of the George A. Smathers Libraries at the University of Florida launched a pilot metadata remediation project using an out-of-box product – Oxygen XML Editor. This article, in addition to providing one more metadata remediation case study, traces the discussions around metadata quality and analyses the general metadata remediation process. Moreover, this article enriches the discussion of vocabulary control in relation to a core function of digital collection sites–faceted search.