KEYNOTE SPEAKERS

John Domingue

Email: j (dot) b (dot) domingue (at) open (dot) ac (dot) uk
Organization: The Knowledge Media Institute, The Open University, http://kmi.open.ac.uk/
Phone: +44 (0)1908 655014
Homepage: http://kmi.open.ac.uk/people/domingue/

John Domingue is the Deputy Director of the Knowledge Media Institute at The Open University and the President of STI International, a semantics focused networking organization with just under 50 members. He has published over 150 refereed articles in the areas of Artificial Intelligence and the Web and his current work is focused on how semantic technology can automate the management, development and use of Web services. Currently he serves as the Scientific Director of SOA4All a 13M Euro project which aims at creating a Web of billions of services. Within the Future Internet arena he helps coordinate the Future Internet Service Offer Working Group within the Future Internet Assembly – a collaboration of 100 European projects with a combined budget of over 600M Euros aiming to develop a next-generation Internet. Prof. Domingue is also serves on the editorial boards for the Journal of Web Semantics and the Applied Ontology Journal.

Keynote talk - Semantic Web Services as a Core Component of the Future Internet

Globally, several major initiatives are currently underway to re-design the Internet meet the economic and societal demands of the 21st Century. Japan, USA and Europe are investing heavily in this area. The EU is shaping around the idea of the Future Internet (www.future-internet.eu) for its research programmes for the Seventh Framework Programme and beyond based upon:

Internet of Things, where every electronic device will be an active participant in the network;
Internet of Services, where applications live in the network, and data becomes an active entity;
Internet of Content and Media, including new technologies for media content provision and where most of the content is generated by end-users.

Following discussions within the EU Future Internet Assemblies (e.g. see www.fi-stockholm.eu and www.r2sconference.eu/sideEvents_fia.php ) and from the emerging results of EU projects such as SOA4All (www.soa4all.eu) it is apparent that a key part of the next generation global network infrastructure will rely on semantic and Web service technologies. In this talk I will outline the main role that semantic and Web service technologies will play giving examples from a number of running EU projects.

Thore Graepel

Email: thoreg (at) microsoft (dot) com
Organization: Microsoft Research, Cambridge, UK http://research.microsoft.com/en-us/
Phone: +44 (0)1223 479 759
Fax: +44 (0)1223 479 999
Homepage: http://research.microsoft.com/en-us/people/thoreg/

Thore Graepel is a senior researcher at Microsoft Research Cambridge (MSRC), UK, and heads the Online Services and Advertising (OSA) research group. The OSA group conducts research in the area of applied machine learning with applications to online advertising, online gaming, and web search. A common theme of Thore's research is the idea of using the tremendous wealth of data Microsoft is generating through its services to improve those services based on data-driven computational intelligence. Thore's research has impacted Microsoft products and services, e.g. through the TrueSkill ranking and matchmaking system in Xbox Live and Halo 3, as well as through his work with adCenter on the prediction of user behaviour. Before starting the OSA group together with Ralf Herbrich, Thore co-founded the Applied Games group at MSRC whose research is aimed at bringing adaptable artificial intelligence to computer games. Thore has a strong academic track record with over 50 peer-reviewed publications in machine learning and probabilistic modelling. His basic research focuses on inference in large scale probabilistic models and knowledge bases. He is also interested in classification and recommender systems, and is passionate about creating a computer program that plays the game of Go better than the best human players.

Keynote talk - Learning to Play: Machine Learning and Computer Games

In the Applied Games group at Microsoft Research Cambridge we are working on the application of machine learning to computer games. I will talk about the motivation for this work and describe three projects in more detail. In the Drivatars project our team built the driving AI for the commercial racing game Forza Motorsport for Xbox. What sets this AI system apart is its ability to learn the driving style of each individual gamer through the application of machine learning methods and provide a "cloned" AI driver that mimics the player's style in game. In the Tao Feng project, we applied the paradigm of reinforcement learning to train an AI system to be able to compete in a fighting game. The system learns to fight by playing against the built-in AI and by adapting its policy from observed state-action-reward sequences. Finally, in the TrueSkill project, we addressed the problem of estimating the skills of gamers in particular games or game modes to be able to match them into balanced, fair, and fun online matches. TrueSkill is based on a principled Bayesian modelling approach and can be viewed as a generalization to the Chess rating system ELO to the multi-player/multi-team scenario.

Atanas Kiryakov

Email: naso.at.sirma.bg
Organization: Ontotext AD, Sirma Group, 135 Tsarigradsko Chausse, Sofia 1784, Bulgaria, http://www.ontotext.com/
Phone: (+359 2) 8091 555
Fax: (+359 2) 8091 426

Atanas Kiryakov is the founder and CEO of Ontotext AD. Ontotext is a leading provider of core semantic technology, distinctive for its performance, scale, and compliance with open standards. Kiryakov joined Sirma as a software engineer in 1993 and later on became partner and member of the board. In year 2000 he found Ontotext lab. He led Ontotext in research projects with total budget above 100 MEuro that allowed it to invest more than 150 man-years in product development. In 2008 Kiryakov finalized an investment deal for Ontotext for 2.5M EURO, under which the VC fund NEVEQ acquired a minority share.

Keynote talk - FactForge – the Fast Track to the Centre of the Data Web
or How reason-able views bring the Semantic Web closer to its tipping point

"Linked data" is a principle for publishing RDF data which allows those to be interlinked and traversed across servers – this way forming a Web of data. LOD is a W3C SWEO Community Project, which aims to facilitate the emergence of a Web of linked data by means of publishing and interlinking open data. Currently LOD includes more than 50 datasets which contain more than 10 billion statements.

Despite the enthusiasm about them the usage of linked data in real-world applications and in the enterprises is still quite low. The main challenges for the adoption of the linked data are that: LOD is hard to comprehend and queried as structured data; it is generally unreliable and inconsistent; querying data distributed across several servers on the Web is slow.

The so-called reason-able views make it easier to use part of the LOD data for specific purposes. Technically to build a reasonable-view one should clean up and integrate selected datasets and ontologies in a compound dataset, which is loaded in a single semantic repository.

FactForge is a reason-able view which includes 8 of the central LOD datasets with total size 1.2 billion explicit. OWL Horst reasoning is performed to "materialize" 0.8 billion facts that could be inferred from these data, while the total number of retrievable statements is 10 billion. FactForge is the only system which provides a solution to the Modigliani test, defined at ReadWriteWeb as the tipping point of the Semantic Web.