Enabling the Web of (Linked Open) Data
Faculty of Science and Technology, University of Westminster, UK
Epaminondas is a Senior Lecturer in Computer Science, Faculty of Science and Technology, University of Westminster, London, UK. He is holding a PhD from the Swiss Federal Institute of Technology (ETH) Zurich, Dept. of Computer Science, Institute of Information Systems, Zurich, Switzerland, as well as a Master’s degree from the University of Karlsruhe, Faculty of Computer Science, Institute of Programme Structures and Data Organisation, Karlsruhe, Germany. Epaminondas has gathered business and entrepreneurial experience as a result of Innovate (TSB) UK funded knowledge transfer partnerships and projects between academic research and enterprises. He also published more than 60 peer reviewed papers in journals and conferences, as well as a recently authored book, Taylor and Francis, CRC Press, all contributing to contemporary research questions in semantic and cognitive computing. Epaminondas is currently teaching computer science related subject areas, such as Web Intelligence and Algorithms.
The Web of Data has been envisioned as a global data repository of linked open data (LOD). Despite many technological advances, which are driven primarily by W3C and the Semantic Web, such as RDF, SPARQL, the main challenge for enabling the Web of Data remains the inherent data heterogeneity, which humbles reusability and linking of open data on the Web. This is mainly due to various challenges of the nature of the problem such as
- difficulty in identifying semantically similar data and data sets to link to or re-use, which are already defined else where by other data providers,
- the diversity of contents on the LOD comprising data about geographic locations, people, genes, proteins, drugs, etc.,
- much of the data is being generated by wrappers around existing relational databases or APIs and, therefore, first need to be crawled before it can be counted or analysed,
- different communities have specific preferences on the vocabularies they prefer to use for publishing data on the Web and, therefore, the Web of Data is open to arbitrary vocabularies being used in parallel.
In order to overcome these challenges and enable successful publishing and linking of open data, the LOD cloud needs to be explored and searched about similar data sets or described entities, which is not a trivial task within such a highly dynamic environment of the Web of Data.
In this tutorial, we will take a look at the Web of Data (LOD project) via the lenses of “emergent semantics”, which is quintessential for a much sought after semantic interoperability in dynamic environments, as well as self-organisation, which is an essential property of emergent semantics systems. The tutorial will stretch over the following two main topics:
- Principles of emergent semantics, e.g., semantic handshaking protocol, evolution from local interactions and agreements towards global ones, and semantic self-organisation, with emphasis on examples found in science and nature, e.g., magnetisation in Physics, or examples from biology and chemistry including the stripped patterns in Zebras, Fish and the ocular dominance columns of the brain. These patterns are produced due to the individual responses of the cells to local conditions and the response of the neighbouring cells.
- Principles of pattern search based information retrieval with the focus on how these apply to the highly dynamic environment of the Web of (Linked Open) Data for the sake of data reusability and interoperability. Particular emphasis will be given on patterns among high-level features in order to bridge the semantic gap in search.
The Web of Data for E-Commerce: Schema.org and GoodRelations for Researchers and Practitioners
E-business and General Management, Universität der Bundeswehr Munich, Germany
Martin Hepp is a professor of E-business and General Management at the Universität der Bundeswehr Munich. He holds a master’s degree in business management and business information systems and a PhD in business information systems from the University of Würzburg (Germany). His key research interests are shared data structures at Web scale, for example Web ontology engineering, both at the technical, social, and economical levels, conceptual modeling in general, and data quality management. As part of his research, he developed the GoodRelations vocabulary, an OWL DL ontology for data interoperability for e-commerce at Web Scale. Since 11/2012, GoodRelations is the e-commerce core of schema.org, the official data markup standard of major search engines, namely Google, Yahoo, Bing, and Yandex. He authored more than 60 academic publications and was the organizer of more than fifteen workshops and conference tracks on conceptual modeling, Semantic Web topics, and information systems, and a member of more than sixty conference and workshop program committees, including ECIS, EKAW, ESWC, IEEE CEC/EEE, ISWC, and WWW.
Schema.org is one of the main drivers for the adoption of needs. GoodRelations is a well-established conceptual model for representing e-commerce information, one of the few widely used OWL DL ontologies, and since 2012 the official e-commerce model of schema.org. In this tutorial, we will (1) give a comprehensive overview and hands-on training on the advanced conceptual structures of schema.org for e-commerce, including patterns for ownership and demand, (2) present the full tool chain for producing and consuming respective data, (3) explain the long-term vision of Linked Open Commerce, and (4) discuss advanced topics, like access control, identity and authentication (e.g. with WebID); micropayment services, and data management issues from the publisher and consumer perspective. We will also cover research opportunities resulting from the growing adoption and the respective amount of data in RDFa, Microdata, and JSON-LD syntaxes.