PAPYRUS is expected to deliver the following results:
- The PAPYRUS semantic multimedia analysis components, which can operate on the content in order to semantically annotate it with concepts of the content (news) domain ontology. These components will be in the form of software for extracting the content structure from news archives and relevant information from secondary sources. They involve the Audio Analysis Component, the Semantic Video Analysis Component, the Contempt Context-based Relevance Feedback Component, the Concept Mapper, the Named Entities Annotator and the PAPYRUS sub-topic annotator
- The PAPYRUS Ontology Tools, which provide all the operations for building the two domain ontologies, for defining the semantic correspondences between them and for semantically interpreting the user queries according to the user (history) domain ontology. These components include:
- The Ontology Development Component (the TRENTO Dataspaces - TRENDS) for ontology editing and mapping, incorporating knowledge mapping tools, which will be software prototypes for the ontology mapping and matching tools between the domains of history and news
- The Ontology Browsing Component for browsing ontologies and submitting queries in the ontology domain, incorporating query analysis and execution module, which will transform the query provided by the user to a multilingual query that will be used to access the multiple news repositories, as well as presentation techniques, which will be attributed in presenting the results to the end users in a meaningful and productive way
- The domain Ontologies for the content (news) and users (history)
- The PAPYRUS Digital Library Engine, which integrate all the provided PAPYRUS functionality. The Search Engine is a core module, which the Digital Library Engine builds upon. Any standard search engine can be used for this purpose. In the current implementation the open-source LUCENE SOLR search engine has been deployed
|
|
|
|