]> The Web-based Systems Group at Freie Universität Berlin explores technical and economic questions concerning the development of global, decentralized information environments. Its current focus lies on the publication and interlinking of structured data on the Web using Semantic Web technologies. The group has initialized several widely used open source software projects including D2RQ, D2R Server, RAP – RDF API for PHP, and NG4G – Named Graphs API for Jena. The group contributes to several open data publishing efforts including DBpedia and the W3C Linking-Open-Data project which aims at interlinking large numbers of data sources on the Web. The Web-based Systems Group is active within the World Wide Web Consortium where it has contributed to the SPARQL recommendation and participates in the Semantic Web Education and Outreach activity. The group maintains strong research ties with the Massachusetts Institute of Technology, Hewlett-Packard Labs, and the Open Archives Initiative. Freie University Berlin will bring in expertise, tools and outreach capabilities to LOD2: (1) FUB has developed and maintains the Silk – Link Discovery Framework and Link Quality Assurance Workbench, which will be significantly extended and integrated into the LOD2 Stack. (2) FUB has developed D2R Server, the most widely used tool for publishing relational databases as Linked Data on the Web. D2R Server will be used for the domain complementation task in WP3 and will be included together with Pubby and Silk into the LOD2 Stack (WP6). (3) Within WP10 Training, Dissemination, community building, FUB will use its existing community building (initiator of W3C LOD) and outreach capabilities (Linked Data on the Web (LDOW) workshop series, Semantic Web Challenge competition series) to maximize the impact of LOD2. 4 WP1 – Requirements, Design and LOD2 Stack Prototype Objectives of this work package are: (1) to develop use case specifications and to collect user requirements by consulting the communities of practice relevant for the LOD2 use cases and additional prospective application scenarios, (2) to identify technical constraints as well as standards, (3) to produce the architecture and the LOD2 Stack design and (4) to produce an early prototype of the LOD2 Stack in the first year. 1 WP5 – Adaptive Linked Data Visualization, Browsing and Authoring The objectives of WP5 are to develop new browsing, visualization and authoring interfaces for LOD, which support a wide range of devices (from mobile phones to desktop PCs), which integrate heterogeneous information from various sources and support the evolution of both instance data as well as information structures over time. In order to achieve these objectives we will explore new browsing and visualization paradigms. 5 WP6 – Interfaces, Component Integration & LOD2 Stack This work package will continue the prototyping activity under WP1/Task 1.4 by fully integrating the individual components developed in WP2-5 into a ready-to-use LOD2 Stack and associated APIs. The primary goal of the LOD2 Stack integration is to enable communities of practice to rapidly create domain specific Linked Data applications. Consequently, the LOD2 Stack will support the whole life cycle of Linked Data from creation over enrichment, interlinking, fusing to maintenance. The stack will be very versatile, for all functionality we will define clear interfaces, which enable the plugging in of alternative third-party implementations. We will also provide a stack configurator, which enables potential user to create their own personalized version of the LOD2 Stack, which contains only those functions relevant for their usage scenario. 6 WP2 – Storing and Querying Very Large Knowledge bases This work package implements the LOD2 knowledge store component needed for managing the Web of Linked Data as a vast database. The starting point is OpenLink Virtuoso and MonetDB on the database side and Sindice on the information retrieval side. The present data volumes handled are around 10 billion RDF triples and the target is over the 1 trillion triples. The approach is scale-out out (physical) complemented with significant scalability improvements in the RDF engine (logical): Physically, when data grows, servers can be added and data redistributed without interruption of service, ibid for server failure. Logically, there is no point answering questions nobody is asking. Therefore the base data is kept as RDF with text indexing and search ranking. Additional inference results or indices for caching joins are made as a by-product of querying; further exploitation of such (partially) materialized inferences through the graph at run-time is to exploit structural correlations in the graphs. 2 WP3 – Knowledge Base Creation, Enrichment and Repair WP3 contains tasks focused on the transformation of legacy data to RDF and Linked Data and furthermore on the improvement of existing or extracted data especially with respect to schema enrichment and ontology repair. It is complementary to WP4, which is concerned with interlinking several knowledge bases and providing unified views of them. Tasks concerning the triplification of data will be grounded on existing techniques and know-how of the consortium and will be refined during the lifetime of this project and integrated into the LOD2 Stack. Legacy data triplification represents the entry point for legacy systems to participate in the LOD cloud. The members of the Consortium are leading in the development of transformational tools such as Virtuoso Sponger, RDF Views, D2R server, Triplify, and the DBpedia framework, which have received high acceptance in the Linked Data community. 3 WP4 – Reuse, Interlinking and Knowledge Fusion While WP3 is concerned with making legacy data available via URLs – a prerequisite – and enrichment of knowledge bases, this WP addresses automatic and semi-automatic link creation with minimal human interaction, evolvement of knowledge bases under the aspect of linkage and schema mapping combined with Data Fusion. 4 WP8 – Use Case 2: LOD2 for Enterprise Data Web This use case will be driven by Exalead, one of the leading enterprise search providers worldwide. We will deploy the LOD2 platform in a real corporate environment with high semantic information integration requirements and needs. Based on the authoring tools and semantic GUIs developed for the LOD2 Stack, we will implement a set of procedures for data input that integrate semantic features and annotations early in the ingestion process of data. We will also suggest a set of best practices for storing, managing, accessing and exchanging data. Moreover, a quality framework will be defined to measure the benefit of using LOD2 components in the corporate. This framework will consist of a set of quality measures to define the gain in precision, recall of searching and browsing the corporate data. Other measurements will be investigated like the impact of the LOD2 platform in the activity of the corporation (income, costs, efficiency, etc). 8 WP9 – Use Case 3: LOD2 for Citizen – GovData.eu The purpose of this GovData.eu use case is to increase public access to high-value, machine-readable data sets generated by the European, national as well as regional governments and public administrations. Although this effort will be similar to developments in other parts of the world, for the case of Europe it will be more challenging due to the larger organizational and linguistic diversity and thus represent an ideal application scenario for Linked Data technologies. 9 WP10 – Training, Dissemination, Community Building, Fertilization The general aim of this work package is to establish a worldwide focal point for academic and industry parties interested in contributing to or taking advantage of the novel Linked Data methodologies and components, which will emerge in the project. 10 Realizing the vision of LOD2 together with the ones of the three use cases will have significant socio-economic impact. Standardization of such an architecture and exploitation of knowledge and technical results (and related IPR) is covered in this work package. 11 The project management will entail strategic, project-wide as well as day-to-day central management and coordination activities. The several different management boards which will be established in the consortium will be responsible for decisions and activities of different scope and level according to their function. 12 D2R Server is a tool for publishing the content of relational databases on the <a href="http://www.w3.org/2001/sw/">Semantic Web</a>, a global information space consisting of <a href="http://www.w3.org/DesignIssues/LinkedData.html">linked data</a>. Data on the Semantic Web is modelled and represented in <a href="http://en.wikipedia.org/wiki/Resource_Description_Framework">RDF</a>. D2R Server uses a customizable <a href="http://www4.wiwiss.fu-berlin.de/bizer/d2rq/spec/#specification">D2RQ mapping</a> to map database content into this format, and allows the RDF data to be browsed and searched – the two main access paradigms to the Semantic Web. D2R Server's Linked Data interface makes RDF descriptions of individual resources available over the HTTP protocol. An RDF description can be retrieved simply by accessing the resource's URI over the Web. Using a Semantic Web browser like <a href="http://www.w3.org/2005/ajar/tab">Tabulator</a> (<a href="http://www.w3.org/2006/Talks/1019-tab-tbl/">slides</a>) or <a href="http://www4.wiwiss.fu-berlin.de/bizer/ng4j/disco/">Disco</a>, you can follow links from one resource to the next, surfing the Web of Data. The SPARQL interface enables applications to search and query the database using the <a href="http://www.w3.org/TR/rdf-sparql-query/">SPARQL</a> query language over the SPARQL protocol. A traditional HTML interface offers access to the familiar Web browsers. <img src="http://www4.wiwiss.fu-berlin.de/bizer/d2r-server/images/architecture.png" alt="D2R Server architecture diagram" /> Requests from the Web are rewritten into SQL queries via the mapping. This on-the-fly translation allows publishing of RDF from large live databases and eliminates the need for replicating the data into a dedicated RDF triple store. Read more about the interfaces offered by D2R Server, including example HTTP requests and responses, in the Technical Note <a href="publishing/">Publishing Databases on the Semantic Web</a>. D2R Server is a tool for publishing the content of relational databases on the <a href="http://www.w3.org/2001/sw/">Semantic Web</a>, a global information space consisting of <a href="http://www.w3.org/DesignIssues/LinkedData.html">linked data</a>. Data on the Semantic Web is modelled and represented in <a href="http://en.wikipedia.org/wiki/Resource_Description_Framework">RDF</a>. D2R Server uses a customizable <a href="http://www4.wiwiss.fu-berlin.de/bizer/d2rq/spec/#specification">D2RQ mapping</a> to map database content into this format, and allows the RDF data to be browsed and searched – the two main access paradigms to the Semantic Web. D2R Server's Linked Data interface makes RDF descriptions of individual resources available over the HTTP protocol. An RDF description can be retrieved simply by accessing the resource's URI over the Web. Using a Semantic Web browser like <a href="http://www.w3.org/2005/ajar/tab">Tabulator</a> (<a href="http://www.w3.org/2006/Talks/1019-tab-tbl/">slides</a>) or <a href="http://www4.wiwiss.fu-berlin.de/bizer/ng4j/disco/">Disco</a>, you can follow links from one resource to the next, surfing the Web of Data. The SPARQL interface enables applications to search and query the database using the <a href="http://www.w3.org/TR/rdf-sparql-query/">SPARQL</a> query language over the SPARQL protocol. A traditional HTML interface offers access to the familiar Web browsers. <img src="http://www4.wiwiss.fu-berlin.de/bizer/d2r-server/images/architecture.png" alt="D2R Server architecture diagram" /> Requests from the Web are rewritten into SQL queries via the mapping. This on-the-fly translation allows publishing of RDF from large live databases and eliminates the need for replicating the data into a dedicated RDF triple store. Read more about the interfaces offered by D2R Server, including example HTTP requests and responses, in the Technical Note <a href="publishing/">Publishing Databases on the Semantic Web</a>. D2R Server is a tool for publishing relational databases on the Semantic Web. It enables RDF and HTML browsers to navigate the content of the database, and allows applications to query the database using the SPARQL query language. DBpedia is a community effort to extract structured information from Wikipedia and to make this information available on the Web. It currently already contains a tremendous amount of valuable knowledge extracted from Wikipedia. The DBpedia knowledge base will be used for evaluation LOD2’s interlinking, fusing, aggregation and visualization components. The DBpedia multi-domain ontology will be used as background-knowledge for the LOD2 applications (WP7, WP8 and WP9), and as an alignment and annotation ontology for LOD in general. DBpedia is a community effort to extract structured information from Wikipedia and to make this information available on the Web. It currently already contains a tremendous amount of valuable knowledge extracted from Wikipedia. The DBpedia knowledge base will be used for evaluation LOD2’s interlinking, fusing, aggregation and visualization components. The DBpedia multi-domain ontology will be used as background-knowledge for the LOD2 applications (WP7, WP8 and WP9), and as an alignment and annotation ontology for LOD in general. SemMF is a flexible framework for calculating semantic similarity between objects that are represented as arbitrary RDF graphs. The framework allows taxonomic and non-taxonomic concept matching techniques to be applied to selected object properties. Moreover, new concept matchers are easily integrated into SemMF by implementing a simple interface, thus making it applicable in a wide range of different use case scenarios. Framework Architecture: he Matching Engine takes as input a query object and a collection of resource objects to be matched against the query object. Both are represented in RDF and may utilize different schema vocabularies. If they use concepts from a common taxonomy, an RDFS or OWL representation of this taxonomy has to be provided. SemMF Engine is implemented in Java utilizing <a href="http://jena.sourceforge.net/">Jena2 Semantic Web Framework</a> for accessing and querying of resource and query graphs as well as the underlying taxonomies. <img src="http://semmf.ag-nbi.de/doc/img/semmf_overview.jpg"> Matching Description: In most cases not every object property is relevant for the similarity computation. For example, an object representing a certain product may contain manufacturer's phone number which may be irrelevant for comparing product's characteristics with customer's preferences (query object). Thus, each relevant property in the query RDF graph must be explicitly specified and mapped to the semantically corresponding property (i.e. holding the same kind of information, e.g. price information) in a resource RDF graph. Each mapping is assigned a <a href="http://semmf.ag-nbi.de/doc/matchers.html">concept matcher</a> implementing a certain matching technique. In a <a href="http://semmf.ag-nbi.de/doc/creatingMD.html">matching description</a> the importance of each object property can be specified by assigning it a certain weight . Moreover, properties to be matched can be grouped into thematic clusters, for example all properties describing technical specification of a product. The property clustering enables to sort the matching result by cluster similarities. The matching description is represented in RDF using <a href="http://semmf.ag-nbi.de/vocabulary/1.1/semmf.rdfs">SemMF vocabulary</a> provided with the framework. Matching Process: Inside each thematic cluster the Engine calculates the similarity between each query property and the corresponding resource property. These similarities are multiplied by the indicated weights and summed up yielding the cluster similarity. All cluster similarities, in turn, multiplied by the specified cluster weights yield the object similarity. However, if for a given query property value there is more than one semantically corresponding resource property value (e.g. a product may be available in different colors) the Engine chooses the one with the highest similarity. Matching Result:: The output of the Matching Engine is a ranking of objects by their similarity values. The Engine also provides a <a href="http://semmf.ag-nbi.de/doc/traversingMatchingResult.html">detailed description of the matching process</a> (i.e. object property values, similarity values for all clusters and for each single object property within a cluster, associated weights, etc.) which can be used to generate explanations for the calculated object similarity. SemMF is a flexible framework for calculating semantic similarity between objects that are represented as arbitrary RDF graphs. The framework allows taxonomic and non-taxonomic concept matching techniques to be applied to selected object properties. Moreover, new concept matchers are easily integrated into SemMF by implementing a simple interface, thus making it applicable in a wide range of different use case scenarios The Web of Data is built upon two simple ideas: First, to employ the RDF data model to publish structured data on the Web. Second, to set explicit <a href="http://www4.wiwiss.fu-berlin.de/bizer/pub/LinkedDataTutorial/#links">RDF links</a> between data items within different data sources. Background information about the Web of Data is found at the wiki pages of the <a href="http://esw.w3.org/topic/SweoIG/TaskForces/CommunityProjects/LinkingOpenData">W3C Linking Open Data community effort</a>, in the overview article <a href="http://tomheath.com/papers/bizer-heath-berners-lee-ijswis-linked-data.pdf">Linked Data - The Story So Far</a> and in the tutorial on <a href="http://www4.wiwiss.fu-berlin.de/bizer/pub/LinkedDataTutorial/">How to publish Linked Data on the Web</a>. The Silk Link Discovery Framework supports data publishers in accomplishing the second task. Using the declarative Silk - Link Specification Language (Silk-LSL), developers can specify which types of RDF links should be discovered between data sources as well as which conditions data items must fulfill in order to be interlinked. These link conditions may combine various similarity metrics and can take the graph around a data item into account, which is addressed using an RDF path language. Silk accesses the data sources that should be interlinked via the SPARQL protocol and can thus be used against local as well as remote SPARQL endpoints. The main features of the Silk link discovery engine are: <ul> <li>Open source link discovery framework, running on all major platforms</li> <li>Support of RDF link generation (owl:sameAs links as well as other types)</li> <li>Flexible, declarative language for specifying link conditions</li> <li>Employment in distributed environments (by accessing local and remote SPARQL endpoints)</li> <li>Usable in situations where terms from different vocabularies are mixed and where no consistent RDFS or OWL schemata exist</li> <li>Scalability and high performance through efficient data handling (speedup factor of 20 compared to Silk 0.2): <ul> <li>Reduction of network load by caching and reusing of SPARQL result sets</li> <li>Multi-threaded computation of the data item comparisons (3 million comparisons per minute on a Core2 Duo)</li> <li>Optional blocking of data items</li> </ul> </li> </ul> Silk is implemented in Scala running on the Java Virtual Machine. In order to run Silk, developers need to: <ol> <li>Have SPARQL access to the datasets that should be interlinked.</li> <li>Write a link specification as described in the <a href="http://www4.wiwiss.fu-berlin.de/bizer/silk/spec/">Silk - User Manual</a>. </li> <li>Install the Silk framework as described in the <a href="http://www4.wiwiss.fu-berlin.de/bizer/silk/spec/index.htm#usage">Installation and Usage</a> section of the manual.</li> </ol> The Silk Linking Framework supports data publishers in setting explicit RDF links between data items within different data sources. Using the declarative Silk - Link Specification Language (Silk-LSL), developers can specify which types of RDF links should be discovered between data sources as well as which conditions data items must fulfil in order to be interlinked. These link conditions may combine various similarity metrics and can take the graph around a data item into account, which is addressed using an RDF path language. The WIQA - Information Quality Assessment Framework is a set of software components for filtering information from the Web using a wide range of different filtering policies. The framework has been designed to fulfill the following requirements: <ul> <li> Flexible Representation of Information together with Quality-Related Meta-information. Information quality assessment may rely on a wide range of different quality indicators. Which quality indicators are relevant depends on the application domain and the quality dimensions to be assessed. Important quality indicators in the context of web-based information systems are provenance information, ratings, and background information about information providers. The WIQA framework uses Named Graphs [<a href="http://www4.wiwiss.fu-berlin.de/bizer/wiqa/#CaBiHaSt04">CaBiHaSt05</a>] as a flexible data model for representing information together with quality related meta-information.</li> <li>Support for different Information Filtering Policies. The relevancy of different quality dimensions and the metrics used to assess these dimensions depend on the application domain, the quality indicators available, the task at hand and the subjective preferences of the information consumer. Therefore, information consumers use a wide range of different information filtering policies in different situations. The WIQA framework allows different policies to be employed for filtering information. Policies are expressed using a declarative policy language and can combine context-, content- and rating-based assessment metrics.</li> <li>Explaining Filtering Decisions. The accuracy of assessment results is often uncertain due to the limited availability of quality indicators and the uncertain quality of the quality indicators themself. Therefore, the final subjective decision of an information consumer whether to trust or distrust assessment results depends on his understanding of the quality indicators and the assessment metrics that have been used in the assessment process. In order to support information consumers in their trust decision, the WIQA framework can generate detailed explanations about filtering decisions.</li> </ul> The Web Information Quality Assessment Framework is a set of software components that empowers information consumers to employ a wide range of different information quality assessment policies to filter information from the Web. Information providers on the Web have different levels of knowledge, different views of the world and different intensions. Thus, provided information may be wrong, biased, inconsistent or outdated. Before information from the Web is used to accomplish a specific task, its quality should be assessed according to task-specific criteria. 2012-04-30 D3.2.3 – Extension of DBpedia Framework 2012-11-30 D4.3.1 – State-of-the-Art Report for Mapping, QA and Data Fusion 2011-01-31 D4.3.2 – Conceptual Design and Implementation of Metrics 2012-02-29 D4.3.3 – Release of the data quality assessment tool 2012-08-31 D4.4.1 – Specification of the mapping publication and discovery framework 2011-06-30 D4.4.2 – Implementation of the mapping publication and discovery framework 2012-02-29 D4.4.3 – Integration of the mapping publication and discovery framework 2013-02-28 D4.5.1 – Initial release of Data Fusion Component 2012-08-31 D4.5.2 – Final release of Data Fusion Component 2013-08-31 D9.4.1 – Report about the availability of eGovernment Datasets 2012-08-31 D10.1.4 – LOD2 PhD workshop and summer school 2013-10-31 National University of Ireland, Galway (NUI, Galway) - Digital Enterprise Research Institute (DERI) is one of the main actors in research and development of semantic technologies in the world. NUIG performs research in the Semantic Web, social networks, sensor network platforms and applies its research results to solve integration problems in various application-oriented projects in eLearning, eGovernment, eBusiness, and eHealth. NUIG develops advanced Semantic Web infrastructures, such as Semantically Interlinked Online Communities (SIOC), semantic search engines (SWSE, Sindice), and platforms for running large-scale, data-intensive experiments, which facilitate collaborative social working environments, scalable storage and reasoning engines, distributed computing, and ontology development. NUIG actively participates in and leads research funded by the EU FP7 program (FAST, Romulus, Okkam, CONET, PECES, iMP), the EU FP6 program (DIP, SUPER, SemanticGov, NEPOMUK, TripCom, RIDE), Science Foundation Ireland (LION) and Enterprise Ireland (SAOR, eLITE). NUIG will guarantee technical excellence in reliable large-scale data processing with the same practices which have been daily driving the works behind the Sindice and Sig.ma projects. NUIG will provide the relevance, feasibility and consensus of the initiative thanks to the continuous interaction between the Linked Data community and the Linked Data Research Centre, a cross institute initiative. 3 OpenLink Software is a United Kingdom SME founded in 1992. It operates a business development and sales subsidiary in the United States. Most product development takes place within the EU, across the UK, Netherlands and Bulgaria. The company is a leading provider of high-performance, scalable, and secure technology covering: data access drivers, data virtualization middleware; native database management (RDBMS or RDF based Graph Store), and enterprise collaboration. Respective product portfolio offerings include: OpenLink High-Performance Drivers for ODBC, JDBC, ADO.NET, OLE-DB, and XMLA; Virtuoso Universal Server; and OpenLink Data Spaces for socially enhanced personal and/or enterprise collaboration. OpenLink has extensive experience in scalable RDF triple (and quad) stores as a result of extending its native Virtuoso database/SQL engine to incorporate SPARQL query language support. Virtuoso also includes powerful Linked Data Deployment and RDF to RDBMS transformation functionality. It offers management and creation of physical and virtual triples in conjunctions with the ability to declaratively produce RDF views of SQL Data (SQL to RDF mapping). OpenLink possess pioneering experience at both the applications and data management levels within the Semantic Web technology realm. OpenLink is a W3C member, an active participant in the W3C Semantic Web Education and Outreach (SWEO) Interest Group, a key member of the Linking Open Data project, and timeless supporter of the Open Data Movement. OpenLink is a DBpedia project co-creator and has hosted live instances of the DBpedia database since project inception. As a consortium member, OpenLink will leverage its experience with local and distributed query processing, SQL and SPARQL query optimization, RDF store and heterogeneous data integration technologies en route to developing an integrated database backbone for the LOD2 project. OpenLink contributes in particular to developing the scalable LOD2 knowledge store (WP2); to track and inference about data provenance and reliability; to support personalized views on knowledge and spatial data; alerts on data; to contribute to standardization activities regarding the integration of semantic and spatial technologies. 5