The first two ODBIS VLDB Workshops were co-located with VLDB 2005 and VLDB 2006 conferences. The joint workshop SWDB-ODBIS was co-located with VLDB 2007. They were all successful with quite rich discussions between participants and good quality papers. LNCS series editors accepted to publish post-proceedings in two LNCS 2007 and LNCS 2008 volumes. This year, we propose a new edition with slightly different objectives: the workshop will focus mainly on Ontology-driven design and management of databases for information systems and knowledge systems.
Indeed ontologies serve as a means for establishing a conceptually concise basis for communicating knowledge in any context. Ontologies can be very useful for a community as a way of structuring and defining the meaning of the metadata that are currently collected and standardized. Applications using ontologies become more intelligent since they can more deal with human background knowledge. More generally, ontologies are critical for applications which require to merge information from diverse sources. They become a major conceptual backbone for a broad spectrum of activities dealing with databases either in information systems or in knowledge systems.
IS professionals and researchers have traditionally dealt with issues of identifying, capturing, and representing domain knowledge within information systems. In the structural dimension, ontologies can provide mechanisms for organizing and storing items including database schemes, user interface objects, and application programs. In the temporal dimension, ontologies can guide the development of new information systems by helping analysts and designers to choose appropriate processes, algorithms, rules, and software components depending upon their needs. The “ontology-driven information systems” approach proposes new ways of thinking about ontologies and IS in conjunction with each other and covers both the structural and the temporal dimensions of information systems.
A key point in databases is the ability to make data available semantically, that is, to find an automated and meaningful way of expressing their structure and semantics. Indeed schemes as sets of rules represent complex agreements made by designers with domain experts about data and so constitute a potentially valuable basic resource for eliciting ontologies. For instance, relational schemes are purely lexical, often obtained from more conceptual ones which are flattened into tables with a loss of information about roles and concepts. Within this perspective, an approach is to search for a tool which will automatically create ontologies corresponding to the content of the database and make them available for humans and machines.
On the other hand, due to their independence from lower level models, ontologies are used for integrating heterogeneous databases enabling interoperability and specifying interfaces to knowledge systems. Availability of the background knowledge stored in ontologies increases significantly the support which can be given for indexing as well as for searching. Ontologies may be useful too for conducting extraction tasks for discovering patterns, interpreting rules or conceptual clustering. Furthermore ontologies can be used to provide semantic annotations for collections of images, audio or other non-textual objects.
The objective of this workshop is to present researches on databases in information systems and knowledge systems as they relate to ontologies and more broadly, to gain insight into ontologies as they relate to databases. It is meant to cover foundations, methodologies and applications of ontology-driven design, integration and management of databases in information systems and knowledge systems.