IC3K is a joint conference composed of three concurrent conferences: KDIR, KEOD and KMIS. These three conferences are always co-located and held in parallel. Keynote lectures are plenary sessions and can be attended by all IC3K participants.
No Smartness Without Data
Dieter A. Fensel, University Innsbruck, Austria
Mining Satellite Images for Census Data Collection: A Study Using the Google Static Maps Service
Frans Coenen, University of Liverpool, United Kingdom
STEALTH: Modeling Coevolutionary Dynamics of Tax Evasion and Auditing
Una-May O'Reilly, MIT Computer Science and Artificial Intelligence Laboratory, United States
From Open Data to Knowledge: Capitalizing Experiences
Marijn Janssen, Delft University of Technology, Netherlands
No Smartness Without Data
Dieter A. Fensel
University Innsbruck
Austria
![]()
Brief Bio
In 1989, Prof. Dr. Dieter Fensel earned both his Master in Social Science (Free University of Berlin) and his Master in Computer Science (Technical University of Berlin). In 1993, he was awarded his Doctoral degree in Economic Science, Dr. rer. pol., from the University of Karlsruhe. And in 1998 he received his Habilitation in Applied Computer Science. Throughout his doctoral and post-doctoral career, Prof. Dr. Fensel has held positions at the University of Karlsruhe (AIFB), the University of Amsterdam (UvA), and the Vrije Universiteit Amsterdam (VU). In 2002, he took a chair at the Institute for Computer Science, Leopold Franzens University of Innsbruck, Austria. In 2003, he became the Scientific Director of the Digital Enterprise Research Institute (DERI) at the National University of Ireland, Galway, receiving a large grant acquired from Science Foundation Ireland (SFI) and in 2006 he became the Director of the Digital Enterprise Research Institute (DERI) at the Leopold Franzens University of Innsbruck, Austria. In 2007, he founded the Semantic Technology Institute International (STI2), which is organized as a collaborative association of interested scientific, industrial and governmental parties of the world wide Semantic Web and Service community that share a common vision. End of 2007, DERI Innsbruck was renamed to STI Innsbruck. STI International already counts 30 associate members from all over the world. His current research interests are around the usage of semantics in 21st century computer science.
He has published over 200 papers via scientific books and journals, conferences, and workshop contributions. He has co-organized over 200 academic workshops and conferences. He is an Associated Editor fourteen scientific journals/publications. He has been an executive member in than 50 international and national research projects with a total volume of more than 40 Million Euro. Furthermore, he worked as scientific or project coordinator in several projects, such as Larck (IP), SOA4All (IP), KnowledgeWeb (NoE) or Tripcom (Strep). His academic experience is not however restricted to academia, having taught over fifty courses at various levels of education, from professional academies and technical colleges to universities and scientific conferences. Topics include: Formal Specification Languages, Software Engineering, Data Warehouse, World Wide Web, Electronic Commerce, Agent-based Information Access, Semantic Web and Ontologies. He has supervised over 50 master theses and PhDs and is a recipient of the Carl-Adam-Petri-Award of the Faculty of Economic Sciences from the University of Karlsruhe (2000). Dieter Fensel has contributed to more than 10 books as an author or editor.
Abstract
AI started with strong expectations 60 years ago.
However, the knowledge acquisition bottleneck started an ice time called the AI winter.
Meanwhile the situation drastically changed.
Large volumes of data and their smart integration provide completely new possibilities. In the talk, we show how assistant systems of Amazon, Apple, Bing, Google and others make usage of semantic technologies to implement a new service layer on top of the current Web. We take the touristic industry as a vertical for illustrating that e-marketing and e-commerce can no longer be done successfully if ignoring these recent trends. Integration semantic technologies in their service provisioning turns from a “nice-to-have” into a “must-have” for the future of such business fields.
Mining Satellite Images for Census Data Collection: A Study Using the Google Static Maps Service
Frans Coenen
University of Liverpool
United Kingdom
![]()
Brief Bio
Prof Frans Coenen is a Professor of Computer Science at The University of Liverpool, UK. He has 36 years of experience of working in AI, ranging from early work on driverless ships to current work on deep and machine learning. He is particularly interested in the application of the techniques of Machine Learning to unusual data sets, such as: (i) graphs and social networks, (ii) time series, (iii) free text of all kinds, (iv) 2D and 3D images, particularly medical images, and (v) video data. He is also interested in data mining over encrypted data. He currently leads a small research group working on many aspects of Machine Learning and AI. He has some 450 refereed research publications, and has been on the programme committees for many Machine Learning and AI conferences and related events. He has supervised over 100 post-graduate research students and postdoctoral research associates, has a substantial research funding track record, and significant experience of working with industry.
Abstract
Census collection is a common practice throughout the world. However, the process is expensive and resource intensive. This is especially the case in areas which feature poor communication and transportation networks. A cost effective alternative is to use high-resolution satellite imagery to obtain a census approximation at a significantly reduced cost. This can be achieved by building a predictor that can label households, that feature in satellite image data, according to “family” size. The fundamental idea is to segment satellite images so as to obtain satellite sub-images describing individual households and to represent these segmentations in a manner conducive to household “family” size prediction. A number of representations are considered: graph-based, histogram based and texture based. By pairing each represented household with known census data, namely family size, a predictor can be constructed to predict household size according to the nature of each representation. The generated predictor can then be used to provide a quick and easy mechanism for the approximate collection of census data that does not require significant resource. To generate the desired predictor training data was obtained by collecting “on ground” census data and matching this to satellite imagery. The test site for the work was a collection of villages lying in the Ethiopian hinterland. The operation of the proposed predictor was evaluated using test data collected in the same manner as the training data, and by utilizing the predictor in the context of a "large scale" study for an area of the Ethiopian hinterland for which the population had been previously recorded.
STEALTH: Modeling Coevolutionary Dynamics of Tax Evasion and Auditing
Una-May O'Reilly
MIT Computer Science and Artificial Intelligence Laboratory
United States
![]()
Brief Bio
Una-May O'Reilly is founder and co-leader of the AnyScale Learning For All (ALFA) group at Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory. ALFA focuses on scalable machine learning, evolutionary algorithms, and frameworks for large scale knowledge mining, prediction and analytics. The group has projects in clinical medicine knowledge discovery, wind energy and MOOC technology. She received the EvoStar Award for Outstanding Achievements in Evolutionary Computation in Europe in 2013. She is a Junior Fellow (elected before age 40) of the International Society of Genetic and Evolutionary Computation, now ACM Sig-EVO. She now serves as Vice-Chair of ACM SigEVO. She served as chair of the largest international Evolutionary Computation Conference, GECCO, in 2005. She has served on the GECCO business committee, co-led the 2006 and 2009 Genetic Programming: Theory to Practice Workshops and co-chaired EuroGP, the largest conference devoted to Genetic Programming. In 2013 she inaugurated the Women in Evolutionary Computation group at GECCO. She is the area editor for Data Analytics and Knowledge Discovery for Genetic Programming and Evolvable Machines (Kluwer), and editor for Evolutionary Computation (MIT Press), and action editor for the Journal of Machine Learning Research.
Abstract
STEALTH is an AI system that detects tax law non-compliance by modeling the co-evolution of tax evasion schemes and their discovery through abstracted audits. Tax evasion accounts for billions of lost income each year. When the government pursues a tax evasion scheme and changes the tax law or audit procedures, the tax evasion schemes evolve and change into an undetectable form. The arms race between tax evasion schemes with tax authority actions presents a significant challenge to guide and plan enforcement efforts.
Acknowledgement: Work done with Jacob Rosen, Erik Hemberg of ALFA (http://groups.csail.mit.edu/ALFA), plus Geoff Warner and Sanith Wijesinghe of MITRE Corporation (http://csail.mit.edu/).
From Open Data to Knowledge: Capitalizing Experiences
Marijn Janssen
Delft University of Technology
Netherlands
www.tbm.tudelft.nl/marijnj
Brief Bio
Prof.dr.ir. Marijn Janssen holds the Antoni van Leeuwenhoek chair in "ICT and Governance" and is head of the Information and Communication Technology section of the Technology, Policy and Management Faculty of Delft University of Technology. Marijn Janssen is co-editor-in-chief of Government Information Quarterly (GIQ), associated editor of International Journal of E-business Research (IJEBR), international Journal of E-Government Research (IJEGR), Decision Support Systems (DSS) and on the editorial board of Information Systems Frontiers (ISF), Transforming Government: People, Process & Policy (TGPPP) and Information Polity (IP). Currently he is involved in the H2020 OpenGovIntelligence project which aims at the automation of the statistical analysis of open data. He is also involved in the H2020 VRE4EIC (A Europe-wide Interoperable Virtual Research Environment to Empower Multidisciplinary Research Communities and Accelerate Innovation and Collaboration) project which improved information sharing among researchers by taking privacy, security and trust issues into account. Dr. Janssen acts as the leading Expert for the European Union in the revision of European Interoperability Framework (EIF) related to the opening of data and data sharing.
Marijn Janssen was ranked as one of the leading e-government and open government data researchers in a survey in 2009 and 2014 and published over 350 refereed publications. More information: www.tbm.tudelft.nl/marijnj.
Abstract
More and more data is opened and shared among public and private organizations. Despite the data deluge the creation of knowledge from open data proves to be more cumbersome. Indiscriminately opening and sharing of data has often limited value, as the gap between open data and the creation of knowledge is not bridged. New tools and instruments are necessary to share valuable insights created from the processing and use of the data. Various initiatives at the technical and organizational level are attempting to bridge this gap. In this keynote the open data vision is revisited, insight from experiences are capitalized and challenges and research directions are presented.