Automatic document-level semantic metadata annotation using folksonomies and domain ontologies

Document indexing for automatic semantic annotation support

Nowadays, capturing the knowledge in ontological structures is one of the primary focuses of the knowledge management research. To exploit the knowledge from the vast quantity of existing unstructured texts available in natural languages in ontologies, tools for automatic semantic annotation (ASA) are heavily needed. In this paper, we present an approach to ASA and a method for documents conten...

Automatic Extraction of Ontologies from Teaching Document Metadata

SITS (Scrutable Intelligent Teaching System) is designed to make use of existing learning items in flexible and effective learning interactions. The reuse of pre-existing resources is important since creating new learning resources is a time consuming task which requires a skilled author. The Internet also provides a large number of resources for reuse. A major hurdle for reuse is metadata, esp...

Document Annotation Support Using Biomedical Ontologies

Adding annotations to documents by extracting data from text yields richer document representation, which users can exploit for various tasks such as search and browsing. However, data extraction is hard, especially in large-scale heterogeneous settings. A more focused technique for data extraction is entity linking, which does not extract new data from documents, but links words in documents t...

Creating Structure from Disorder - Using Folksonomies to Create Semantic Metadata

This paper reports on an on-going research project to create educational semantic metadata out of folksonomies. The paper describes a simple scenario for the usage of the generated semantic metadata in teaching, and describes the ‘FolksAnnotation’ tool which applies an organization scheme to tags in a specific domain of interest. The contribution of this paper is to describe an evaluation frame...

Semantic-Based Image Retrial in the VQ Compressed Domain using Image Annotation Statistical Models