Conference Paper

ACoAR: a method for the automatic classification of annotated resources.

DOI: 10.1145/1597735.1597772 Conference: Proceedings of the 5th International Conference on Knowledge Capture (K-CAP 2009), September 1-4, 2009, Redondo Beach, California, USA
Source: DBLP


We propose a classification method that automatically classifies annotated resources under the concepts of a classification system represented by an ontology. We use two well known systems used to classify web pages, for the folksonomy information and DMOZ for an existing ontology, to validate the method. Results obtained provide a correct classification rate of resources of 78%, rising to 93% when using an adequate threshold.

1 Read
  • Source
    • "Since its first release in 2004 the W3C recommendation SKOS (Simple Knowledge Organisation System) has been utilized by several semantic web applications as a lightweight model to support interoperability at the terminological and schematic level (See [3], [6], [5]). Its comparably low ontological (semantic) complexity makes SKOS an ideal standard to be utilized for collaborative knowledge organization purposes especially within the context of socially generated classification schemes (i.e. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents conceptual assumptions about the interaction between the structural specificities of a thesaurus and the quality of a thesaurus-based application output. So far hardly any literature exists that discusses thesaurus modelling requirements with respect to the following thesaurus-specific application areas: classifying, indexing, autocom-plete, query expansion, recommendation and glossaries. By looking at these application areas the authors compare the structural attributes of SKOS and discuss their functional relevance. The authors conclude that taking these assumptions into account can significantly support application-oriented thesaurus modelling hence incrementally improving thesaurus-based applications in terms of modelling scope and effort. An empirical testing of these assumptions is subject to future work.
    Proceedings the 7th International Conference on Semantic Systems, I-SEMANTICS 2011, Graz, Austria, September 7-9, 2011; 01/2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents an automatic method to group resources of collaborative-social tagging systems in semantic categories. The main goal is to self-adapt the system to represent the current knowledge.
    Proceedings of the 6th International Conference on Knowledge Capture (K-CAP 2011), June 26-29, 2011, Banff, Alberta, Canada; 01/2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: Nowadays, folksonomies are currently the simplest way to classify information inWeb 2.0. However, such folksonomies increase continuously their amount of information without any centralized control, complicating the knowledge representation. We analyse a method to group resources of collaborative-social tagging systems in semantic categories. It is able to automatically create the classification categories to represent the current knowledge and to self-adapt to the changes of the folksonomies, classifying the resources under categories and creating/deleting them. As opposed to current proposals that require the re-evaluation of the whole folksonomy to maintain updated the categories, our method is an incremental aggregation technique which guarantees its adaptation to highly dynamic systems without requiring a full reassessment of the folksonomy.
    7th International Conference on Knowledge Management in Organizations: Service and Cloud Computing; 07/2012

Similar Publications