c 2018 for the individual papers by the papers' authors, unless indicated otherwise. Copying permitted for private and academic purposes. Re-publication of material from this volume requires permission by the copyright owners, unless indicated otherwise. matthias.wauer@uni-leipzig.de GeoLD 2018 Geospatial data is vital for many application scenarios, such as navigation, logistics, and tourism. At the same time, a large number of currently available datasets (both RDF and conventional) contain geospatial information. Examples include DBpedia, Wikidata, Geonames, OpenStreetMap and its RDF counterpart, LinkedGeoData. RDF stores have become robust and scalable enough to support volumes of billions of records (RDF triples). Likewise, geospatial information systems (GIS) can bene t from Linked Data principles (e.g., schema agility and interoperability).
Despite improving implementations and standards such as GeoSPARQL, traditional geospatial data management systems still have advantages in functionality, e ciency and scalability regarding geospatial content. In this third edition of the workshop on Geospatial Linked Data, about 20 participants have discussed the current state of GeoLD tools, applications and novel research.
After a short introduction by the organizing committee chair, Ali Khalili presented a convincing approach to a concept called Functional Urban Areas. Several tools, also targeted towards end-users, can be used for integrating Linked Data within spatial boundaries extracted from di erent datasets. He showed how this combination of data can provide new insights that aren't captured by the data provided by the OECD1. The only weakness of this paper is a lack of related work, resulting in a best workshop paper nomination.
Alan Meehan then discussed an approach for ne-grained access control on geospatial data. He argued that a combination of template and licence, described by novel RDF vocabularies, would provide the necessary restrictions for the analysed use cases. Although the paper contains a section on related work, the authors do not compare against these or comment on another approach for ne-grained access control in Apache Rya, as suggested by a reviewer.
After a co ee break, Finn Arup Nielsen presented an extension to their website visualising scienti c data from Wikidata. While the presented use cases and maps were interesting for the audience, reviewers argued that the method of how the queries are generated is not made clear.
In the following short paper presentation, Peru Bhardwaj explained the issues she ran into when executing link discovery. This talk was controversial and caused a lively discussion with the audience. While some improvement suggestions should be addressed by the link discovery tool developers, it remained unclear why the application of unsupervised or active learning approaches was not successful.
Finally, Matthias Wauer presented an early version of a platform for integration geospatial and sensor data. Using semantic technologies and a message bus based on RabbitMQ, the approach and implementation were motivated by three use cases. The discussion with the audience led to the suggestion of comparing 1Organization for Economic Co-operation and Development the approach with the Linked Data Noti cations W3C recommendation which could be relevant.
To summarize, the workshop discussed many aspects of geospatial linked data, perhaps with the notable exception of data quality. An audience member concluded that now that there are several full-featured GeoSPARQL-capable stores and other tools available, it will be interesting to see which applications will be possible in recent years. The importance of the topic was further highlighted by several presentations in the "Semantic Geo Resources" session of the main conference.
We thank the authors for their submissions and the program committee for their hard work.
June 2018
Matthias Wauer, Mohamed Sherif, Axel-Cyrille Ngonga Ngomo
Michael Martin, Institute for Applied Informatics Anisa Rula, Fraunhofer IAIS Giorgos Giannopoulos, Athena Institute for the Management of Information Systems Kleanthi Georgala, University of Leipzig Roman Korf, USU Software AG Harsh Thakkar, Fraunhofer IAIS Muhammad Saleem, University of Leipzig Richard Wacker, YellowMap AG Tim Ermilov, University of Leipzig Henning Hasemann, TomTom Ricardo Usbeck, University of Paderborn Konstantina Bereta, National and Kapodistrian University of Athens Ivan Ermilov, University of Leipzig Johannes Trame, metaphacts QuWeDa 2018 The constant growth of Linked Open Data (LOD) on the Web opens new challenges pertaining to querying such massive amounts of publicly available data. LOD datasets are available through various interfaces, such as data dumps, SPARQL endpoints and triple pattern fragments. In addition, various sources produce streaming data. E ciently querying these sources is of central importance for the scalability of Linked Data and Semantic Web technologies. The trend of publicly available and interconnected data is shifting the focus of Web technologies towards new paradigms of Linked Data querying. To exploit the massive amount of LOD data to its full potential, users should be able to query and combine this data easily and e ectively. This workshop at the Extended Semantic Web Conference (ESWC) presented original articles describing theoretical and practical methods and techniques for fostering, querying, and consuming the Data Web. The workshop brought together members of the community interested in demonstrating their latest advances in query processing systems for RDF. The event fostered discussion for proposing novel RDF query processing techniques, language extension, and benchmarking and experimental evaluation of the engines.
June 2018
Olaf Hartig, Linkoping University Ricardo Usbeck, Universitat Paderborn Ruben Verborgh, Ghent University Axel-Cyrille Ngonga Ngomo, Universitat Paderborn
Stasinos Konstantopoulos, NCSR Demokritos Maribel Acosta, Karlsruhe Institute of Technology Aidan Hogan, DCC, Universidad de Chile Harald Sack, FIZ Karlsruhe, Leibniz Institute for Info. Infr. and KIT Karlsruhe Pascal Molli, University of Nantes - LS2N Ste en Staab, WeST, Univ. Koblenz-Landau and WAIS, Univ. of Southampton Peter Haase, metaphacts Andreas Schwarte, uid Operations AG Soren Auer, TIB Leibniz Information Center Science+Tech. / Univ. Hannover Alessandro Adamou, The Open University Gong Cheng, Nanjing University Giuseppe Pirro, Institute for High Performance Computing and Networking Andriy Nikolov, metaphacts GmbH Danh Le Phuoc, TU Berlin Oscar Corcho, Universidad Politecnica de Madrid Katja Hose, Aalborg University Stefan Dietze, L3S Research Center Carlos Buil Aranda, Universidad Tecnica Federico Santa Mar a Luis Ibanez-Gonzalez, University of Southampton Enrico Daga, The Open University Hala Skaf-Molli, Nantes University Stefan Schlobach, Vrije Universiteit Amsterdam Valeria Fionda, University of Calabria Vanessa Lopez, IBM Ali Hasnain, Insight Centre for Data Analytics Olivier Corby, INRIA Using Linked Open Geo Boundaries for Adaptive Delineation of Functional Urban Areas
Ali Khalili, Peter van den Besselaar, and Klaas Andries de Graaf License and Template Access Control for Geospatial Linked Data Alan Meehan, Kaniz Fatema, Rob Brennan Eamonn Clinton, Lorraine McNerney, and Declan O'Sullivan Geospatial data and Scholia
Finn Arup Nielsen, Daniel Mietchen, and Egon Willighagen On the Overlooked Challenges of Link Discovery
Peru Bhardwaj, Christophe Debruyne, and Declan O'Sullivan Towards a Semantic Message-driven Microservice Platform for Geospatial and Sensor Data
Matthias Wauer and Axel-Cyrille Ngonga Ngomo Benchmarking Commercial RDF Stores with Publications O ce Dataset Ghislain Auguste Atemezing and Florence Amardeilh 9 22 34 41 47 59 Heuristics-based Query Reordering for Federated Queries in SPARQL 1.1 and SPARQL-LD Thanos Yannakis, Pavlos Fafalios and Yannis Tzitzikas 74