Collection-Level User Searches in Federated Digital Resource Environment

Abstract

As part of a federation project providing integrated access to over 160 digital collections, we are studying how collections can best be represented to meet the needs of diverse user communities. This paper reports preliminary transaction log analysis results from that project on subject representation of the digital collections. The findings reveal prevalence of the broadly defined subject search at the collection level, and the lack of semantic similarity between the user queries and the GEM controlled vocabulary terms used for collection description. Based on the actual search data, it is recommended that the 2nd group of entities in the FRBR model be updated to cover ethnic/national groups and classes of persons. The paper proposes definitions of the two major collection-level search types — known-item and subject — and formulates research questions for further investigation into subject access to federated collections.