2009 | OriginalPaper | Buchkapitel
Tags4Tags: Using Tagging to Consolidate Tags
verfasst von : Leyla Jael Garcia-Castro, Martin Hepp, Alexander Garcia
Erschienen in: Database and Expert Systems Applications
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Tagging has become increasingly popular and useful across various social networks and applications. It allows users to classify and organize resources for improving the retrieval performance over those tagged resources. Within social networks, tags can also facilitate the interaction between members of the community,
e.g.
because similar tags may represent similar interests. Although obviously useful for straightforward retrieval tasks, the current meta-data model underlying typical tagging systems does not fully exploit the potential of the social process of finding, establishing, challenging, and promoting symbols,
i.e.
tags. For instance, the social process is not used for establishing an explicit hierarchy of tags or for the collective detection of equivalencies, synonyms, morphological variants, and other useful relationships across tags. This limitation is due to the constraints of the typical meta-model of tagging, in which the subject must be a Web resource, the relationship type is always
hasTag,
and the object must be a tag as a literal. In this paper, we propose a simple yet effective extension for the current meta-model of tagging systems in order to exploit the potential of collective tagging for the emergence of richer semantic structures, in particular for capturing semantic relationships between tags. Our approach expands the range of the object of tagging from Web resources only to the union of (1) Web resources and (2) pairs of tags, i.e., users can now use arbitrary tags for expressing typed relationships between a pair of tags. This allows the user community to establish similarity relations and other types of relationships between tags. We present a first prototype and the results from an evaluation in a small controlled setting.