Elise Conradi to _ be _ classified A Facet

This paper examines the use of the postulational approach to facet analysis to manually induce a faceted classification ontology from a folksonomy. An in-depth study of faceted classification theory is used to form a methodology based on the postulational approach, which is then used to facet analyze a dataset consisting of over 107,000 instances of 1,275 unique tags representing 76 popular nonfiction history books collected from the LibraryThing folksonomy. Preliminary results of the facet analysis indicate the manual inducement of two faceted classification ontologies in the dataset: a completed ontology representing the domain of books and an incomplete ontology representing the domain of subjects within the domain of books. The grouping of tags into theoretically based facets and conceptual categories give new insight into how users describe information resources. Furthermore, the relationships discerned in the ontologies are usergenerated relationships between tagged information items, representing a new form of knowledge. Practical implications of the results are discussed in terms of potential areas in which user-generated metadata can enhance faceted structures in information architecture. FOLKSONOMIES: UNSTRUCTURED “WISDOM OF THE CROWD” Since their inception on the web in 2003 with the tagging system Del.icio.us, folksonomies have become a popular way to categorize large amounts of information resources. Folksonomies emerge from the aggregation of textual labels called tags that are affixed to digital objects of various formats by either the creator or the users of the objects within sites that allow for tagging. Vander Wal (2005) distinguishes between broad folksonomies and narrow folksonomies, explaining that a “broad folksonomy has many people tagging the same object” whereas in a narrow folksonomy, an object is tagged “by one or a few people.” Quintarelli (2005) discusses broad folksonomies in terms of the Power Law distribution, stating that the “power law reveals that many people agree on using a few popular tags but also that smaller groups often prefer less known terms to describe their items of interest.” Halpin et al. (2007) argue that the short head of the long tail in the Power Law distribution represents a consensus of what users find to be most important about each information resource. The study of folksonomies can therefore provide invaluable insight 1 http://del.icio.us JOURNALOFIA.ORG | ISSN 1903-7260 5 ELISE CONRADI | TO_BE_CLASSIFIED to information organization professionals by unearthing what Weinberger (2006) has called the “wisdom of the crowd”. Folksonomies have been criticized by those advocating top-down approaches to organizing information resources. It is argued that the uncontrolled vocabulary of tags causes too many recall and precision problems (primarily due to ambiguity, polysemy and synonymy) to make them useful as information retrieval tools, and that the flat structure of folksonomies prevent users from seeing valuable relationships between information items (Rosenfield, 2005; Petersen, 2006). In response to the latter critique, this paper illustrates how a facet analysis of a broad folksonomy based on the postulational approach can reveal underlying conceptual categories and facets to which the folksonomy’s aggregated tags belong. In this way, facet analysis techniques are used to manually expose a faceted classification ontology in the flat tag space, thus revealing user-generated relationships between information items. RELATED WORK There are several studies and projects that have examined the use of faceted classification techniques for the organization of folksonomies. Weaver (2007) studied the tagging practices of a library community in order to glean facets to aid in information retrieval. Quintarelli, Resmini and Rosati (2007) introduced “Facetag”, a tagging system that allows users to choose tags within predefined facets in order to improve retrieval. Lichtblau, Trice and Wartik (2006) proposed a prototype social classification system, in which users describe services within a specific domain of the Department of Defense from seven different facets based on “the 7 W’s”. Siderean launched the wonderfully named but short-lived site Fac.etio.us in 2005, in which tags were automatically grouped into predefined facets. Other commercial enterprises combining the use of tags with facets are Buzzillions, Peter Van Dijck’s brainchild MeFeedia, and Raw Sugar, a “guided, tag-based search engine”. Spiteri (2010) analyzes several of these attempts and concludes that, although “a number of studies exist in which facets have been applied quite successfully to social tagging applications, (...) none explain clearly the theoretical frameworks or methodologies used to derive the facets, nor do they address any strategies by which to enable end users to evaluate the usefulness and applicability of these facets” (p. 105). Indeed, with the exception of Weaver (2007), all of the above studies are largely focused on the improvement of faceted navigation and information retrieval through the 2 1. Who uses the service? 2. What does the service do? 3. On what does the service act? 4. To whom is the service generally directed? 5. Where is the service used? 6. When is the service used? 7. Why is the service used? 3 Fac.etio.us, by way of the Internet Archive Wayback Machine: http://web.archive.org/web/20060526050202/demo.siderean.com/facetious/facetious.jsp 4 http://www.buzzillions.com 5 http://www.mefeedia.com 6 http://www.rawsugar.com