Navigation auf


Department of Informatics

Details Colloquium Spring 2013

21.02.2013 - BabelNet: a Very Large Multilingual Ontology and its Applications

Speaker: Prof. Dr. Roberto Navigli
Host: Martin Volk


In the information society, lexical knowledge is a key skill for understanding and decoding an ever-changing world. However, lexical knowledge is an essential component not only for human understanding of text, but also for natural language driven tasks such as automated text understanding, information retrieval and question answering. Unfortunately, building such lexical knowledge resources manually is an onerous task requiring dozens of years - and furthermore it has to be repeated from scratch for each new language. I will present a novel integration and enrichment methodology that produces a very large multilingual semantic network: BabelNet ( This resource is created by linking the largest multilingual Web encyclopedia - i.e., Wikipedia - to the most popular computational lexicon of English - i.e., WordNet. The integration is performed via an automatic mapping and by filling in lexical gaps in resource-poor languages with the aid of Machine Translation. The result is an "encyclopedic dictionary" that provides babel synsets, i.e., concepts and named entities lexicalized in many languages and connected with large amounts of semantic relations. In the last part of the talk I will showcase the new BabelNet 1.1 and its API. I will also present the current applications of BabelNet to state-of-the-art multilingual word sense disambiguation and semantic relatedness.


Prof. Dr. Roberto Navigli is an Associate Professor in the Department of Computer Science of the Sapienza University of Rome. He was awarded the Marco Cadoli 2007 Italian National Prize for the best doctoral thesis in Artificial Intelligence and he is the recipient of an ERC Starting Grant 2010 in computer science and informatics, entitled "multilingual joint word sense disambiguation". His research lies in the field of Natural Language Processing (including word sense disambiguation and induction, ontology learning and large-scale knowledge acquisition). Currently he is a member of the editorial board of Computational Linguistics and the Journal of Natural Language Engineering, and a guest editor of the Artificial Intelligence Journal and the Journal of Web Semantics.

07.03.2013 - Inside the Mind of Watson

Speaker: Chris Welty
Host: Avi Bernstein


OWatson is a computer system capable of answering rich natural language questions and estimating its confidence in those answers at a level of the best humans at the task. On Feb 14-16, in an historic event, Watson triumphed over the best human players of all time on the American television quiz-show, Jeopardy! In this talk I will discuss how Watson works at a high level with examples from the show.


Chris Welty Ph.D. is a Research Scientist at the IBM T.J. Watson Research Center in New York. Previously, he taught Computer Science at Vassar College, taught at and received his Ph.D. from Rensselaer Polytechnice Institute, and accumulated over 14 years of teaching experience before moving to industrial research. Chris' principal area of research is Knowledge Representation, specifically ontologies and the semantic web, and he spends most of his time applying this technology to Natural Language Question Answering as a member of the DeepQA/Watson team and, in the past, Software Engineering. Dr. Welty is a co-chair of the W3C Rules Interchange Format Working Group (RIF), serves on the steering committee of the Formal Ontology in Information Systems Conferences, is president of KR.ORG, on the editorial boards of AI Magazine, The Journal of Applied Ontology, and The Journal of Web Semantics, and was an editor in the W3C Web Ontology Working Group. While on sabbatical in 2000, he co-developed the OntoClean methodology with Nicola Guarino. Chris Welty's work on ontologies and ontology methodology has appeared in CACM, and numerous other publications.

18.04.2013 - Human-Centered Interaction and Experience Design of Smart Cities

Speaker: Dr. Dr. Norbert Streitz
Host: Gerd Schwabe


Having entered what is being called the Urban Age, where more than half of the world population is living in cities, economic prosperity and quality of life will largely depend on the abilities of cities to reach their full potential. One important dimension beyond the deployment of appropriate infrastructures is providing ambient intelligence-based support for smart urban living. Urban environments become increasingly interactive spaces reflecting also in the real world the so far mainly in the virtual world created social networks. Combining these information and experience spaces with ubiquitous computing in urban contexts constitutes what is being called "smart hybrid city". The talk addresses issues and challenges for designing ambient intelligence environments, especially from a human-environment interaction perspective. This includes the shift from information design to experience design, spreading social communication behavior from virtual worlds back into real environments. We are arguing for a people-oriented, empowering smartness where smart spaces make people smarter by keeping the human in the loop. The implications of sensor-based smart environments (e.g., availability and use of location-based services) for privacy will be discussed because they reach a new dimension. Privacy will become a commodity people have to pay for and thus a privilege. Needing a vision for reconciling humans and technology in the Urban Age, we argue for a human-centered design approach resulting in a Humane Smart Hybrid City where people can exploit their creative potential and lead a self-determined life.


Dr. Dr. Norbert Streitz is a Senior Scientist and Strategic Advisor with more than 30 years of experience in information and communication technology. He is the founder and scientific director of the Smart Future Initiative (SFI) which was launched in January 2009. From 1987 - 2008, he was at the Fraunhofer Institute IPSI (previously GMD-IPSI) in Darmstadt, Germany, where he held different positions as Deputy Director and Division Manager. At IPSI, he initiated and managed research efforts in different areas (human-computer-interaction, hypertext/hypermedia, CSCW, ubiquitous computing, ambient intelligence). A prominent example is the research division "AMBIENTE - Smart Environments of the Future" founded by him in 1997 and the development of Roomware®. He also taught at the Department of Computer Science of the Technical University Darmstadt for more than 15 years. Before joining IPSI in Darmstadt, he was an Assistant Professor at the Technical University Aachen (RWTH), Germany, teaching and doing research in cognitive science and human-computer interaction and founding the ACCEPT-Group (AaChen Cognitive Ergonomics ProjecT). This was preceded by his work in theoretical physics at the University of Kiel, Germany. Furthermore, he was a post-doc research fellow at the University of California, Berkeley, USA, a visiting scholar at Xerox PARC, USA, and at the Intelligent Systems Lab of MITI, Tsukuba Science City, Japan. He is regularly asked to present keynote speeches and tutorials at scientific as well as commercial events in Europe, USA, South America, Qatar, Malaysia, Singapore, Hongkong, China, Korea and Japan.

25.04.2013 - Delivering the Smart Grid: A Grand Challenge for Autonomous Agents and Multi-agent Systems Research

Speaker: Dr. Alex Rogers
Host: Sven Seuken


Restructuring electricity grids to meet the increased demand of electric vehicles and heat pumps, while making greater use of intermittent renewable energy sources, represents one of the greatest engineering challenges of our day. This modern electricity grid, in which both electricity and information flow in two directions between large numbers of widely distributed suppliers and generators - commonly termed the 'smart grid' - represents a radical re-engineering of infrastructure which has changed little over the last hundred years. However, the autonomous behavior expected of the smart grid, its highly distributed nature, and the existence of multiple stakeholders each with their own incentives and interests, challenges existing engineering approaches. In this talk, I will describe why I believe that autonomous agents and multi-agent systems are essential for delivering the smart grid as it is envisioned. I will present some recent work that has been done in this area, and describe many challenges that still remain.


Dr. Alex Rogers is a Reader in the Agents, Interaction and Complexity group at the University of Southampton. He originally studied Physics, and then worked in the oil industry as a field engineer. After working in various offshore locations around the world, he returned to academia to study for a PhD in statistical modelling of complex system. After a short stint in a spin out from the Santa Fe which was trying to apply these approaches to business problems, he returned to academia for a second time, and his research now focuses on developing and applying agent-based algorithms and mechanisms for the control of decentralized systems. He is particularly interested in decentralized information systems such as sensor networks and future energy networks such as the smart grid. His work lies at the intersection of artificial intelligence, data fusion, game theory and microeconomics.

02.05.2013 - Designing "Smart" User Interfaces for the Smart Grid Market

Speaker: Prof. Dr. Sven Seuken
Host: Martin Glinz


The Internet has allowed electronic markets like eBay or Amazon to become ubiquitous. However, new technologies are continuously enabling new kinds of markets in previously non-market domains, like the energy market for example. While these new markets generally increase users' choice sets which can increase efficiency, they are also often too complex for non-expert users to handle effectively. This gives rise to an interesting research challenge at the intersection of market design and user interface (UI) design. In this talk, I will present three of my research projects that addresses this challenge. First, I will briefly describe a "hidden" peer-to-peer backup market that we designed and implemented, and I will show a demo of the prototype system. Second, I will present results from a lab experiment on automatic market user interface optimization. But for the majority of my talk, I will focus on an ongoing research project on designing "smart" user interfaces for the smart grid market. The particular challenge we are currently looking at is how to design a smart thermostat for a domain with very dynamic energy prices. We developed an active learning algorithm that minimizes the user interactions necessary, only asking for feedback when absolutely necessary. The system learns the user's desired trade-off between comfort and price over time, and thus is able to automatically set the thermostat on the user's behalf. I will close my talk with a short discussion on future challenges in the area of market user interface design for the smart grid.


Prof. Dr. Sven Seuken is an Assistant Professor (Tenure Track) at the Department of Informatics of the University of Zurich and head of the Computation and Economics Research Group. He was a Fulbright Scholar, and the recipient of a Microsoft Research Ph.D. Fellowship, a Harvard Teaching Award (Certicate of Distinction in Teaching), and a Fellowship from the German National Academic Foundation. Seuken received a Ph.D. in Computer Science from Harvard University in 2011, an M.S. degree from Harvard in 2008, an M.S. degree in Computer Science from the University of Massachusetts, Amherst in 2006, and a pre-diploma from the University of Freiburg in 2003. His research lies at the intersection of Computer Science and Economics, with a particular focus on electronic market design.

07.05.2013 - Mobile and Ubiquitous Computing Tools for Learning the Environment

Speaker: Prof. Khai Truong
Host: Elaine Huang


A rich cognitive map of an environment can enhance an individual’s experience within the space. However, acquiring knowledge about an environment can be challenging for people with visual impairments. They often need support by sighted people to understand what is in the environment, associate meaning to these objects and places, identify where different objects and places are positioned, and their spatial relationship with each other. In this talk, I will present the potential role that mobile and ubiquitous computing tools can play in helping a user learn an environment. I will discuss what information is required by visually impaired pedestrians, when and where to provide this information, and also how to collect and present it. Finally, I will describe some of my recent and ongoing work in this application space; I will explain how they can help visually impaired users increase their independence and confidence in their navigation.


Prof. Khai N. Truong, Ph.D., is an Associate Professor in the Department of Computer Science at the University of Toronto. His research lies at the intersection of Human-Computer Interaction (HCI) and Ubiquitous Computing (UbiComp), specifically examining the mutual impact of usability and technical constraints on the design of applications and interaction techniques for novel, off-the-desktop computing systems that may be commonplace in 5-10 years. He received his PhD in computer science from the Georgia Institute of Technology.

23.05.2013 - Towards Smart Computational Design

Speaker: Prof. Dr. Helmut Pottmann
Host: Renato Pajarola


Traditional shape design systems provide the user with little support to satisfy constraints implied by function and fabrication of the designed product. Thus, we see a trend towards research aiming at a much closer connection between shape creation, function and manufacturing. Within this wide and largely unexplored area, the speaker will concentrate on (i) architecture and (ii) geometric approaches. Specifically, we will address tools for the design of architectural freeform structures composed of simple and easily manufacturable elements (panels, support structure) and for exploring the variety of possible designs under given constraints,including aspects of statics. Particular emphasis is put on the geometrization of manufacturing and structural constraints and on partially high-dimensional models for shape exploration and form ending processes. From a mathematical perspective, this work is interesting due to its close relations to discrete dierentialgeometry and the geometry of constraint manifolds and shape spaces.


Prof. Dr. Helmut Pottmann is director of the Geometric Modeling and Scientific Visualization Center at King Abdullah University of Science and Technology in Saudi Arabia. Prior to joining KAUST he has been professor of geometry at Vienna University of Technology and head of the "Geometric Modeling and Industrial Geometry" research group. His research interests are in Applied Geometry and Visual Computing, in particular in Geometric Modeling, Geometry Processing and most recently in Geometric Computing for Architecture and Manufacturing.

Weiterführende Informationen


Teaser text