Hosted at FIS2010, the 3rd Future Internet Symposium. September 20-22, 2010 - Berlin
The Internet of Things and the Web of Data promise a future where everything will be connected and where data can be seamlessly exchanged throughout highly heterogeneous networks. However, it is yet unclear how these incredible amounts of messy data will be dealt with. The goal of this workshop is to bring together researchers from the Internet and Web communities with more heuristic-oriented problem-solving communities, such as the people interested in Computational Intelligence (CI) and Nature-inspired algorithms.
A particular focus will be set on the Web of Data (WoD) for which the billions of facts hosted by many different parties, represented using a variety of vocabularies with varying degrees of preciseness, and supplied in an inconsistent fashion yield a high level of messiness. This WoD highlights the scalability and robustness problems the Future Internet will have to deal with.
The Future Internet considers new approaches to realize an information transport and processing infrastructure. An important component of it, the Web of Data, carries out the vision of a network of information suitable for consumption from both humans and machines. This web, consisting of inter-connected instance data annotated with possibly expressive ontologies, promises a huge opportunity for Web-based applications in many domains. In this Web where everyone is free to create information, centralized top-down approaches to control the data available are expensive and hard to design. Just as for the Web of Documents, any centralized tools have to be approximate: Google does not index all information on the WWW, and actually, its users would not need to known every page matching their requests.
Existing reasoning techniques making use of the WoD often fail to live up to this distributed vision of knowledge on the web. Reasoners may not be able to cope with the high number of instances and/or the expressiveness of the ontologies describing them (scalability problem) or fail to deal with a sometimes unreliable network, lack of maintenance of information or the variety in quality (robustness problem).
The problems of scalability and robustness of reasoning on the Web have been widely recognised in the Semantic Web (SW) community. To deal with them, the urge for alternative reasoning methods has been investigated: e.g. incomplete reasoning techniques are often able to provide some answers faster than complete ones. To cope with the variety on the web, and the distribution of the data and knowledge, the research community looks more and more into alternative reasoning paradigms. More than only thinking of targeted fixes for specific problems, there is a need for the SW research community to focus on returning "good enough" answers. This line of research falls under the heading of approximation, a domain well studied by researchers working on fields such as Computational Intelligence and Nature-inspired optimization techniques.
Nature inspired mechanisms are applicable to further areas of interest in the effort to fulfill the Future Internets promise of location independent, interoperable, coherent, consistent, scalable, pervasive, reliable, secure and efficient access to a coordinated set of services. The expected outcome of this workshop is to reach out to new communities and to open up to the use of new paradigms within the SW community. A short list of keywords for the workshop would be:
On a broader point of view, any heuristic approach providing anytime approximate solutions to a problem are likely to be of interest to the participants of this workshop.
July 26th, 2010 (23:59 pm Hawaii time, GMT-10)
Extended Deadline: August 22nd, 2010 (23:59 pm Hawaii time, GMT-10)
September 5, 2010
September 20, 2010
Papers are requested to be written in English and formatted using the Springer Publications format for Lecture Notes in Computer Science (LNCS) and to contain 4-6 pages. Submissions will be made using the EasyChair Conference System, and proceedings of the papers will be provided through the CEUR online service.
The best papers will be considered for a submission as an extended version to an international journal. For more information, please see http://www.mdpi.com/journal/futureinternet/special_issues/fis2010/
We plan to use the following Workshop schedule and structure: