Gabriele Anderst-Kotsis
Johannes Kepler University Linz, Austria
"Omnipresent Multimedia – Pain and Gain of the Always Connected Paradigm"
Abstract: What we call the world wide web nowadays has its origins in a system designed to share and distribute information among scientists. It took less then a decade and “the web” evolved into an omnipresent medium for information sharing, communication and even collaboration and cooperation. This evolution was accompanied by advances in wireless network technologies, increasing capabilities of mobile devices and ongoing “digitalization” of all types of media. A multitude of new standards, protocols, and services has been proposed, some of them became a success, others died even before they had the chance to lift up. In the first part of this talk, we are going to reflect on the impact of omnipresent multimedia in many application domains. We will argue towards the need for rich media interaction, i.e. going beyond the classical multimedia formats of audio and video. We will highlight the importance of considering QoS in service development and deployment leading to a high level of user satisfaction. In the second part of this talk, we will critically reflect potential downsides of those trends, ranging from well known concerns about security and privacy to problems of information overload and the risk of prefering digital instead of human communication. The talk should end in a discussion on what we as the scientific community in computer science, can and should do to ensure beneficial and sustainable advances in technology.
Short Bio: Gabriele Anderst-Kotsis is holding a full professor position in computer science at Johannes Kepler Universität Linz. She is chairing the Department of Telecooperation with a research focus in mobile computing, multimedia and hypermedia systems as well as cooperative and collaborative systems. Research in those areas includes the investigation of methods, techniques and tools for system development as well as evaluation and analysis with focus on performance evaluation. The Department is participating in numerous national and international projects, including CRUISE, a European network of excellence in sensor networks, EuroFGI, a network of excellence on Future Generation Internet, the AustrianGrid project, or ModelCVS a project on semantics in SW and system modelling, and actively involved in the organisation of international conferences, including for example iiWAS and MoMM. Prof. Anderst-Kotsis is author of numerous publications in international conferences and journals and is co-editor of several books. She is currently Member at large of ACM.
Ioana Manolescu
Senior Researcher, Inria Saclay and Ecole Polytechnique, France
"Structural and Semantic Summarization of RDF Graphs"
Abstract: The Resource Description Framework (RDF) is the W3C’s graph data model for Semantic Web applications. RDF graphs are often large and heterogeneous, thus users may have a hard time getting familiar with the structure and semantics of a graph. We consider the problem of building automatically, with no user input, compact RDF graph summaries which represent the complete structure and semantics of the graph. Specifically, we study RDF graph summaries based on the classical concept of graph quotient. We show that any such summary is representative (has some answer for any query that has answers on the graph) and accurate (any query having answers on the summary has answer on some graph which it summarizes; many graphs may have the same summary). We provide the first technique for building out of an RDF graph a summary of its explicit and implicit data (the latter is due to RDF semantic constraints); a summary for which this is possible is termed complete. We introduce four novel summaries and show that two of them are complete. We provide a sufficient condition for RDF summarization completeness, and show that bisimulation-based summaries previously studied satisfy this condition. We implemented a summarization tool available online and demonstrate its effectiveness through a set of experiments. Joint work with Sejla Cebiric (Inria Saclay) and François Goasdoué (Irisa and U. Rennes 1)
Short Bio: Ioana Manolescu is a senior researcher at Inria Saclay and Ecole Polytechnique. She is the lead of the CEDAR INRIA team focusing on rich data analytics at cloud scale. She is a member of the PVLDB Endowment Board of Trustees, of the ACM SIGMOD Jim Gray PhD dissertation committee, and an associate editor for PVLDB. Recently, she has been the program chair of the Scientific and Statistical Data Management Conference 2016, and she is a co-chair of the upcoming IEEE ICDE conference in 2018. She has co-authored more than 130 articles in international journals and conferences, and contributed recently to a book on "Web Data Management" by S. Abiteboul, I. Manolescu, P. Rigaux, M.-C. Rousset and P. Senellart. Her main research interests algebraic and storage optimizations for semistructured data and in particular data models for the Semantic Web, novel data models and languages for complex data management, data models and algorithms for fact-checking, and distributed architectures for complex large data.
Carlos Ordonez
University of Houston, USA
"Can Parallel Database Systems Help Big Data Analytics?"
Abstract: Big data analytics remains an active research area. We argue parallel database systems can indeed solve important, but not all, big data problems. We present a review of parallel processing alternatives, storage mechanisms, scalable algorithms and data structures. We discuss three major tasks: data preprocessing (integrating, cleaning, aggregating, transforming data), computing common machine learning models (clustering, dimensonality reduction, regression, classification) and solving graph problems (paths, clique detection, page rank). On the mathematical side, we contrast relational tuples/tables and vectors/matrices as fundamental abstractions. On the systems side, we compare features of parallel DBMSs (evaluating SQL queries, UDFs, C/C++ code) and the Hadoop stack (HDFS, MapReduce/Spark, Java/Scala code) as the two major competing technologies. We conclude with an outline of solved problems and open research issues. This talk should be interesting to both practitioners and researchers.
Short Bio: Carlos Ordonez studied at UNAM, Mexico, getting a B.Sc. in applied math and an M.S. in computer science. He continued PhD studies at the Georgia Institute of Technology focusing on extending database systems with data mining algorithms. Thereafter Carlos worked at NCR Corp. for 8 years, collaborating in the optimization of machine learning and cube algorithms to work on the Teradata parallel DBMS. In 2006 Carlos joined the University of Houston, where he currently leads the Parallel Database Systems lab. From 2013 to 2015 Carlos collaborated with Michael Stonebraker, regularly visiting the Database Group at MIT. From July 2014 to July 2015 Carlos was a visiting researcher at ATT Labs (formerly ATT Bell Labs), where he worked on stream analytics and big data warehousing with Divesh Srivastava. Carlos has served as Editor on the IEEE TKDE, Intelligent Data Analysis and Information Systems journals and has been on the PC of venues like SIGMOD, KDD, DOLAP, ICDE, CIKM, DaWaK, and AMW. His research has been funded by NSF.
Roland Traunmüller
University of Linz, Austria
"Digitalization as Challenge for Electronic Government"
Abstract: Digitalization is a megatrend that conveys key challenges for e-Government and touches several hot topics. Knowledge is the fundamental bridge bringing together diverse extensions. Important developments include: Administrative Work, Collaborative Platforms, Open Government Data, e-Participation, Mobile Government etc.
Short Bio: Roland Traunmüller is Professor Emeritus with the Institute for Informatics in Business and Government at Linz University, Austria. Prof. Traunmüller has worked in the field of Information Systems and applications of information technology in Government for three decades. In Austria Prof. Traunmüller heads the Forum e-Government, in Germany he is member of the steering body e-Government within the German Computer Society. Within DEXA Prof. Traunmüller founded in 2002 the EGOV conference series as annual meeting of the European R&D Community. In addition he has been involved in various consulting activities and boards on the national and international level (Ministries, EU, UNO, and UNESCO). In recognition of his work official acknowledgements have come in. In recognition of his work official acknowledgements have come in. So for founding the e-Government R&D Community he got 2006 the "Prometheus" - an Award dedicated by the North-American Society on Digital Government and by the European Community for e-Government. He also got the Roland Wagner Award - an international award on computers helping people with special need. Lately the Republic of Austria honoured the Life-work of Prof. Traunmüller with awarding to him the Big Silber Medal of the Republic Austria.