This paper puts forward a MARC-based semiautomated approach to extracting semantically rich subject facets from general and/or specialized controlled vocabularies for display in topic-oriented faceted catalog interfaces in a way that would better support users' exploratory search tasks.
Hierarchical faceted subject metadata is extracted from general and/or specialized controlled vocabularies by using standard client/server communication protocols. Rigorous facet analysis, classification and linguistic principles are applied on top of that to ensure faceting accuracy and consistency.
A shallow application of facet analysis and classification, together with poorly organized displays, is one of the major barriers to effective faceted navigation in library, archive and museum catalogs.
This paper does not deal with Web-scale discovery services.
This paper offers suggestions that can be used by the technical services departments of libraries, archives and museums in designing and developing more powerful exploratory search interfaces.
This paper addresses the problem of deriving clearly delineated topical facets from existing metadata for display in a user-friendly, high-level topical overview that is meant to encourage a multidimensional exploration of local collections as well as “learning by browsing.”
