Lynda Hardman, Centrum Wiskunde & Informatica and University of Utrecht, Netherlands
From Linked Data to Stories: Using explicit semantics to improve interactive information access
People have used narrative to communicate since before the emergence of written historical records. Narrative continues to form the basis of modern forms of video-based communication, such as feature films, documentaries and news items.
Users of video search engines, however, have to deduce the relations among the information contained in the clips to infer a coherent story from them. This leads to unnecessary cognitive effort that does not directly satisfy the viewer’s information need that led to the query. Viewers would be better served if a selection of relevant video clips were assembled into a coherent sequence that conforms to a familiar communication structure.
To assemble a selection of related video clips into a coherent video sequence requires information about the content of the presentation and, additionally, about the communication structure, such as narrative or argument. The Linked Open Data cloud is a rich source of domain information and, in conjunction with annotations of media fragments, can be used to select related pieces of content. Information of how to assemble these fragments into a coherent whole is required.
We illustrate the construction of coherent video sequences based on argumentation structure, and explore what other communication structures may be possible. Examples are drawn from better contextualising news events (LinkedTV News), constructing a coherent video documentary and assessing the implications of a shale-gas drilling on the environment.
Slides available here.
Axel Polleres, Institute for Information Business, WU Wien, Austria
(Enterprise) Linked Data: What’s missing?
In this talk, I will discuss gaps and challenges to deploy the ideas of Linked Data and the Semantic Web, from an Enterprise point of view. Shifting away from a predominantly academic perspective and an exclusive focus on open data, many of us conceive Linked Data as an emerging key disruptive technology for enterprise data management. Still, various challenges lie ahead to lay the foundations for successful Enterprise Linked Data product and services development.
In a current research project, PROPEL: Propelling the potential of Enterprise Linked Data in Austria, we gauge industry requirements and innovation potentials as well as technology & standards gaps for Linked Data to be effectively usable in enterprises. We'll report on the findings and on potential contributions to fill these gaps from our own research at WU Wien.
Slides available here.
Steffen Staab, Institute WeST - Web Science and Technologies & Institute for Computer Science, University of Koblenz-Landau, Germany
Programming with Semantic Broad Data*
The challenges of Big Data are frequently explained by dealing with Volume, Velocity, Variety and Veracity. The large variety of data in organizations results from accessing different information systems with heterogeneous schemata or ontologies. In this talk I will present the research efforts that target the management of such broad data. They include: (i) an integrated development environment for programming with broad data, (ii) a query language that allows for typing of query results, (iii) a typed lambda-calculus based on description logics, and (iv) efficient access to data repositories via schema indices.
Filip Železný, Department of Computer Science, FEE CTU in Prague, Czech Republic
Semantic web and machine learning: time to re-sync
I will review some recent developments due to different authors in the intersection of knowledge representation and semantic web on one hand, and machine learning on the other hand. The talk will address the state of the art in symbolic learning using logical frameworks native to the semantic web community, but it will also cover some interesting results triggered by the statistical and deep-learning stream, such as embedding knowledge bases into tensor or connectionist representations allowing the induction of novel concepts.
*Keynote is collocated with the Prague Computer Science Seminar.