The Semantic Web: ESWC 2012 Satellite Events

Despite the recent growth in the size of the Linked Data Cloud, the absence of links between the vocabularies of the sources has resulted in heterogenous schemas. Our previous work tried to find conceptual mapping between two sources and was successful in finding alignments, such as equivalence and subset relations, using the instances that are linked as equal. By using existential concepts and their intersections to define specialized classes (

restriction classes

), we were able to find alignments where previously existing concepts in one source did not have corresponding equivalent concepts in the other source. Upon inspection, we found that though we were able to find a good number of alignments, we were unable to completely cover one source with the other. In many cases we observed that even though a larger class could be defined completely by the multiple smaller classes that it subsumed, we were unable to find these alignments because our definition of

restriction classes

did not contain the disjunction operator to define a union of concepts. In this paper we propose a method that discovers alignments such as these, where a (larger) concept of the first source is aligned to the union of the subsumed (smaller) concepts from the other source. We apply this new algorithm to the Geospatial, Biological Classification, and Genetics domains and show that this approach is able to discover numerous concept coverings, where (in most cases) the subsumed classes are disjoint. The resulting alignments are useful for determining the mappings between ontologies, refining existing ontologies, and finding inconsistencies that may indicate that some instances have been erroneously aligned.

The DBpedia project extracts structured information from Wikipedia and makes it available on the web. Information is gathered mainly with the help of infoboxes that contain structured information of the Wikipedia article. A lot of information is only contained in the article body and is not yet included in DBpedia. In this paper we focus on the extraction of historical events from Wikipedia articles that are available for about 2,500 years for different languages. We have extracted about 121,000 events with more than 325,000 links to DBpedia entities and provide access to this data via a Web API, SPARQL endpoint, Linked Data Interface and in a timeline application.

The ready availability of data is leading to the increased opportunity of their re-use for new applications and for analyses. Most of these data are not necessarily in the format users want, are usually heterogeneous, and highly dynamic, and this necessitates data transformation efforts to re-purpose them. Interactive data transformation (IDT) tools are becoming easily available to lower these barriers to data transformation efforts. This paper describes a principled way to capture data lineage of interactive data transformation processes. We provide a formal model of IDT, its mapping to a provenance representation, and its implementation and validation on Google Refine. Provision of the data transformation process sequences allows assessment of data quality and ensures portability between IDT and other data transformation platforms. The proposed model showed a high level of coverage against a set of requirements used for evaluating systems that provide provenance management solutions.

The increasing availability of data on the Web provided by the emergence of Web 2.0 applications and, more recently by Linked Data, brought additional complexity to data management tasks, where the number of available data sources and their associated heterogeneity drastically increases. In this scenario, where data is reused and repurposed on a new scale, the pattern expressed as Extract-Transform-Load (ETL) emerges as a fundamental and recurrent process for both producers and consumers of data on the Web. In addition to ETL,

provenance

, the representation of source artifacts, processes and agents behind data, becomes another cornerstone element for Web data management, playing a fundamental role in data quality assessment, data semantics and facilitating the reproducibility of data transformation processes. This paper proposes the convergence of these two Web data management concerns, introducing a principled provenance model for ETL processes in the form of a vocabulary based on the Open Provenance Model (OPM) standard and focusing on the provision of an interoperable provenance model for ETL environments. The proposed ETL provenance model is instantiated in a real-world sustainability reporting scenario.

The eXtensible Business Reporting Language (XBRL) has standardised consolidated financial reporting and through its machine readable format facilitates access to and consumption of financial figures contained within the report. Formalising XBRL as RDF facilitates the leveraging of XBRL with Open Financial Data. Previous XBRL to Semantic Web transformations have however concentrated on making the semantics of its logical model explicit to the exclusion of accounting regulatory validation rules and constraints found within the XBRL calculation linkbases. Using off-the-shelf Semantic Web technologies this paper investigates the use of the SPARQL Inferencing Notation (SPIN) with RDF to formalise these accounting regulations found across XBRL jurisdictional taxonomies. Moving beyond previous RDF to XBRL transformations we investigate how SPIN enhanced formalisation enables financial instrument fact inferencing and sophisticated consistency checking. SPIN formalisations are further used to evaluate the correctness of reported financial data against the calculation requirements imposed by accounting regulation. Our approach illustrated through the use of use case demonstrators outlines that SPIN usage meets central requirements for financial constraint regulatory modelling.

Opinion mining is a sub-discipline of computational linguistics that uses information retrieval techniques in order to determine whether a piece of text expresses a positive, negative or neutral opinion. In this paper, we present an approach for the opinion mining of financial news through the process of identifying their semantic polarity. Our approach relies on an algorithm that combines several gazetteer lists and leverages an existing financial ontology. The financial-related news are obtained from RSS feeds and then automatically annotated with positive or negative markers. The outcome of the process is a set of news organized by their degree of positivity and negativity. The preliminary experimental results seem promising as compared against traditional approaches.

Online Analytical Processing (OLAP) promises an interface to analyse Linked Data containing statistics going beyond other interaction paradigms such as follow-your-nose browsers, faceted-search interfaces and query builders. Transforming statistical Linked Data into a star schema to populate a relational database and applying a common OLAP engine do not allow to optimise OLAP queries on RDF or to directly propagate changes of Linked Data sources to clients. Therefore, as a new way to interact with statistics published as Linked Data, we investigate the problem of executing OLAP queries via SPARQL on an RDF store. First, we define projection, slice, dice and roll-up operations on single data cubes published as Linked Data reusing the RDF Data Cube vocabulary and show how a nested set of operations lead to an OLAP query. Second, we show how to transform an OLAP query to a SPARQL query which generates all required tuples from the data cube. In a small experiment, we show the applicability of our OLAP-to-SPARQL mapping in answering a business question in the financial domain.

With the evolution of linked open data sources, question answering regains importance as a way to make data accessible and explorable to the public. The triple structure of RDF-data at the same time seems to predetermine question answering for being devised in its native subject-verb-object form. The devices of natural language, however, often exceed this trFiple-centered model. But RDF does not preclude this point of view. Rather, it depends on the modeling. As part of a government funded research project named Alexandria, we implemented an approach to question answering that enables the user to ask questions in ways that may involve more than binary relations.

Much research has been done to combine the fields of Databases and Natural Language Processing. While many works focus on the problem of deriving a structured query for a given natural language question, the problem of query verbalization – translating a structured query into natural language – is less explored. In this work we describe our approach to verbalizing SPARQL queries in order to create natural language expressions that are readable and understandable by the human day-to-day user. These expressions are helpful when having search engines that generate SPARQL queries for user-provided natural language questions or keywords. Displaying verbalizations of generated queries to a user enables the user to check whether the right question has been understood. While our approach enables verbalization of only a subset of SPARQL 1.1, this subset applies to 90 % of the

$$209$$

queries in our training set. These observations are based on a corpus of SPARQL queries consisting of datasets from the QALD-1 challenge and the ILD2012 challenge.

This paper reports on the first usage of the MultiFarm dataset for evaluating ontology matching systems. This dataset has been designed as a comprehensive benchmark for multilingual ontology matching. In a first set of experiments, we analyze how state-of-the-art matching systems – not particularly designed for the task of multilingual ontology matching – perform on this dataset. These experiments show the hardness of MultiFarm and result in baselines for any algorithm specifically designed for multilingual ontology matching. We continue with a second set of experiments, where we analyze three systems that have been extended with specific strategies to solve the multilingual matching problem. This paper allows us to draw relevant conclusions for both multilingual ontology matching and ontology matching evaluation in general.

Recent work on searching the Semantic Web has yielded a wide range of approaches with respect to the style of input, the underlying search mechanisms and the manner in which results are presented. Each approach has an impact upon the quality of the information retrieved and the user’s experience of the search process. This highlights the need for formalised and consistent evaluation to benchmark the coverage, applicability and usability of existing tools and provide indications of future directions for advancement of the state-of-the-art. In this paper, we describe a comprehensive evaluation methodology which addresses both the underlying performance and the subjective usability of a tool. We present the key outcomes of a recently completed international evaluation campaign which adopted this approach and thus identify a number of new requirements for semantic search tools from both the perspective of the underlying technology as well as the user experience.

The ever-increasing quantity of data produced in biomedical applications requires the development of intelligent tools implementing Knowledge Organization Systems (KOS) like ontologies, thesauri, or classification schemes for a better integration and exploitation of this data. However, due to the size of this application field, element overlapping can occur between different KOS relying on various knowledge representation models. Therefore, mappings are necessary to improve the semantic interoperability of systems using KOS. Moreover, due to the dynamics of the biomedical domain, KOS constantly evolve over time, and new versions are released periodically forcing domain experts to revise the existing mappings affected by the evolution of the underlying heterogeneous KOS. In this paper, we provide an analysis of the mapping evolution problem. We highlight the lacks of existing approaches to deal with this problem and outline the sketch of the DyKOSMap framework as a prospective solution.

There is an increasing need to interconnect biomedical ontologies. We investigate a simple but promising approach to generate mappings between ontologies by reusing and composing existing mappings across intermediate ontologies. Such an approach is especially promising for highly interconnected ontologies such as in the life science domain. There may be many ontologies that can be used for composition so that the problem arises to find the most suitable ones providing the best results. We therefore propose measures and strategies to select the most promising intermediate ontologies for composition. We further discuss advanced composition techniques to create more complete mappings compared to standard mapping composition. Experimental results for matching anatomy ontologies demonstrate the effectiveness of our approaches.

The adoption of ICT technologies in healthcare for recording patients’ health events and progression in Electronic Health Records (EHRs) and Clinical Information Systems (CLIS) has led to a rapidly increasing volume of data which is, in general, distributed in autonomous heterogeneous databases. The secondary use of such data (commonly anonymised for privacy reasons) for purposes other than healthcare (such as patient selection for clinical trials) comprises an emerging trend. However, this trend encapsulates a great challenge; semantic interlinking of two different, yet highly related, domains (in terms of semantics) i.e., clinical research and healthcare. This paper aims at presenting an analysis of the heterogeneity issues met in this effort and describing the semantically-enabled multi-step process followed within the PONTE project for achieving the interlinking of these two domains for the provision of the size of the eligible patients for participation in a trial at the cooperating sites.

Mobile devices are becoming major repositories of personal information. Still, they do not provide a uniform manner to deal with data from both inside and outside the device. Linked data provides a uniform interface to access structured interconnected data over the web. Hence, exposing mobile phone information as linked data would improve the usability of such information. We present an API that provides data access in RDF, both within mobile devices and from the outside world. This API is based on the Android content provider API which is designed to share data across Android applications. Moreover, it introduces a transparent URI dereferencing scheme, exposing content outside of the device. As a consequence, any application may access data as linked data without any a priori knowledge of the data source.

The World Wide Web as it is currently deployed can only be accessed using modern client devices and graphical interfaces, within an infrastructure compassing datacenters and reliable, high-speed Internet connections. However, in many regions in developing countries these conditions are absent. Many people living in remote rural areas in developing countries will not be able to use the Web, unless they can produce and consume voice-based content using alternative interfaces such as (2G) mobile phone, and radio. In this paper we introduce a radio platform, based on a use case and requirements analysis of community radio stations in Mali. The voice-based content of this radio platform will be made publicly available, using Linked Data principles, and will be ready for unexpected re-use. It will help to bring the benefits of the Web to people who are out of reach of computers and the Internet.

In this paper we introduce the Social Media Matrix, a practitioner-oriented instrument that supports companies to decide based on their corporate or communication goals which social media activities to execute. The matrix consists of three parts: 1. Social media goals and task areas have been identified and matched. 2. Five types of social media activities have been defined. 3. The matrix provides a structure to assess the suitability of each activity type on each social media platform for each goal. Whereas the first two parts can be generally used, the assessment must be conducted explicitly for an industry sector. A ready to use assessment for the German B2B sector has been exemplarily compiled from expert-interviews with practitioners and by reviewing concrete social media activities. The matrix is used as a basis for social media consultancy projects and evaluated thereby.

We discuss the challenge of scalable dissemination and communication approaches in a world where the number of channels is growing exponentially. The web, Web 2.0, and semantic channels have provided a multitude of interaction possibilities providing significant potential for yield, brand, and general reputation management. Our goal is to enable smaller organizations to fully exploit this potential. To achieve this, we have developed a new methodology based on distinguishing and explicitly interweaving content and communication as a central means for achieving content reusability, and thereby scalability over various heterogeneous channels.

We describe a system for meta-analysis where a wiki stores numerical data in a simple comma-separated values format and a web service performs the numerical statistical computation. We initially apply the system on multiple meta-analyses of structural neuroimaging data results. The described system allows for mass meta-analysis, e.g., meta-analysis across multiple brain regions and multiple mental disorders providing an overview of important relationships and their uncertainties in a collaborative environment.

In an often retweeted Twitter post, entrepreneur and software architect Inge Henriksen described the relation of Web 1.0 to Web 3.0 as: “

Web 1.0 connected humans with machines. Web 2.0 connected humans with humans. Web 3.0 connects machines with machines

.” On the one hand, an incredible amount of valuable data is described by billions of triples, machine-accessible and interconnected thanks to the promises of Linked Data. On the other hand,

rest

is a scalable, resource-oriented architectural style that, like the Linked Data vision, recognizes the importance of links between resources. Hypermedia

api

s are resources, too—albeit dynamic ones—and unfortunately, neither Linked Data principles, nor the

rest

-implied self-descriptiveness of hypermedia

api

s sufficiently describe them to allow for long-envisioned realizations like automatic service discovery and composition. We argue that describing inter-resource links—similarly to what the Linked Data movement has done for data—is the key to machine-driven consumption of

api

s. In this paper, we explain how the description format

rest

desc captures the functionality of

api

s by explaining the effect of dynamic interactions, effectively complementing the Linked Data vision.

The Linked Data effort has been focusing on how to publish open data sets on the Web, and it has had great results. However, mechanisms for updating linked data sources have been neglected in research. We propose a structure for Linked Data resources in named graphs, connected through hyperlinks and self-described with light metadata, that is a natural match for using standard HTTP methods to implement application-specific (high-level) public update APIs.

This paper describes methodology used for building a domain specific ontology. Methods that allow automatic concept and relation extraction using domain-related glossaries are presented in this research. The constructed ontology contains vocabulary related to computer science and software development. It is used for supporting different functionalities in the ALERT project, which aims to improve software development process in large open source communities. One of the uses of the ontology is to provide semantic search functionality, which is a considerable improvement over the keyword search that is commonly supported.

Nowadays, datasets become larger and larger. As stated by Eric Schmidt, every two days now we create as much or more information as we did from the dawn of civilization up until 2003. Thus, the question is how to make good use of such big data. A solution is Complex Event Processing (CEP) engines that propose to correlate realtime, contextual and past information. In this paper we propose a new architecture that leverages existing research done in Publish/Subscribe systems and CEP engines with the idea to federate them in order to scale to the load that could be encountered in todays ubiquitous events workloads.

This paper refers to methods and tools for enabling context detection and management based on events. We propose a context model that builds on top of previous efforts and we give details about the mechanisms developed for context detection in event marketplaces. In addition, we show how simple or complex events can be used in combination with external services in order to derive higher level context with the use of Situation-Action-Networks (SANs). Specifically, we present two different approaches, one for detecting low level context and another one for deriving higher-level contextual information using SANs. We present an illustrative scenario for demonstrating the process of specialization of our generic context model and its instantiation based on real-time events.

Most of the existing hierarchical Distributed Hash Table (DHT) systems, used for a resource discovery, generate considerable maintenance overhead which affects the routing efficiency in large scale systems. In this paper, we propose a Scalable Distributed Data Structures (SDDS) based Hierarchical DHT (SDDS- HDHT) solution for an efficient data source discovery in data Grid systems. Our solution deals with a reduced number of gateway peers running a DHT protocol. Each of them serves also as a proxy for second level peers in a single Virtual Organization (VO), structured as a SDDS. The performance evaluation of the proposed method proved the discovery cost reduction especially for intra-VO resource discovery queries. It also proved significant system maintenance save especially when peers frequently join/ leave the system.

In distributed software contexts, Web Services (

WSs

) that provide transactional properties are useful to guarantee reliable Transactional Composite

WSs

(

TCWSs

) execution and to ensure the whole system consistent state even in presence of failures. Failures during the execution of a

TCWS

can be repaired by forward or backward recovery processes, according to the component

WSs

transactional properties. In this paper, we present the architecture and an implementation of a framework, called FaCETa, for efficient, fault tolerant, and correct distributed execution of

TCWSs

. FaCETa relies on

WSs

replacement, on a compensation protocol, and on unrolling processes of Colored Petri-Nets to support failures. We implemented FaCETa in a Message Passing Interface (MPI) cluster of PCs in order to analyze and compare the behavior of the recovery techniques and the intrusiveness of the framework.

Sgvizler is a small JavaScript wrapper for visualization of SPARQL results sets. It integrates well with HTML web pages by letting the user specify SPARQL SELECT queries directly into designated HTML elements, which are rendered to contain the specified visualization type on page load or on function call. Sgvizler supports a vast number of visualization types, most notably all of the major charts available in the Google Chart Tools, but also by allowing users to easily modify and extend the set of rendering functions, e.g., specified using direct DOM manipulation or external JavaScript visualization tool-kits. Sgvizler is compatible with all modern web browsers.

This demo presents a web application which implements a pipeline for searching and browsing through newspaper archives. It uses a combination of information extraction, enrichment and visualization algorithms to help users to grasp a large amount of articles normally collected in archives. Illustrative results show appropriateness of the proposed pipeline for searching and browsing news archives.

Most of the CMS platforms lack the management of semantic information about the content although a lot of research has been carried out. The IKS project has introduced a reference architecture for Semantic Content Management Systems (SCMS). The objective is to merge the latest advancements in semantic web technologies with the needs of legacy CMS platforms. Apache Stanbol is a part of this SCMS reference implementation.

Recently practical approaches for managing and supporting the life-cycle of semantic content on the Web of Data made quite some progress. However, the currently least developed aspect of the semantic content life-cycle is the user-friendly manual and semi-automatic creation of rich semantic content.

This demo shows a mobile Android app that uses openly available geographic data and crowdsources parking availability information, in order to let its users conveniently find parking when coming to work or driving into town. The application builds on Linked Data, and publishes the crowdsourced parking availability data openly as well. Further, it integrates additional related data sources, such as events and services, to provide rich value-adding features that will act as an incentive for users to adopt the app.

While it is easy to find statistics on almost every topic, coming up with an explanation about those statistics is a much more difficult task. This demo showcases the prototype tool

Explain-a-LOD

, which uses background knowledge from DBpedia for generating possible explanations for a statistic (This demo accompanies the ESWC paper

Generating Possible Interpretations for Statistics from Linked Open Data

[

1

].).

The ScienceWISE system is a collaborative ontology editor and paper annotation tool designed to help researchers in their discovery. In this paper, we describe the system currently deployed at

sciencewise.info

and the exposition of its data as Linked Data. During the “RDFization” process, we faced issues to encode the knowledge base in SKOS and find resources to link to on the LOD. We discuss these issues and the remaining open challenges to implement some target features.

The Large Knowledge Collider (LarKC) is a prominent development platform for the Semantic Web reasoning applications. Guided by the preliminary goal to facilitate the incomplete reasoning, LarKC has evolved in a unique platform, which can be used for the development of robust, flexible, and efficient semantic web applications, also leveraging the modern grid and cloud resources. As a reaction on the numerous requests coming from the tremendously increasing user community of LarKC, we set up a demonstration package for LarKC that is intended to present the main subsystems, development tools and graphical user interfaces of LarKC. The demo aims for both early adopters and experienced users and serves the purpose of promoting Semantic Web Reasoning and LarKC technologies to the potentially new user communities.

The application of methodologies for building ontologies can improve ontology quality. However, such quality is not guaranteed because of the difficulties involved in ontology modelling. These difficulties are related to the inclusion of anomalies or bad practices within the ontology development. Several authors have provided lists of typical anomalies detected in ontologies during the last decade. In this context, our aim in this paper is to describe OOPS! (OntOlogy Pitfall Scanner!), a tool for detecting pitfalls in ontologies.

The ICE-Map Visualization was developed to graphically analyze the distribution of indexing results within a given Knowledge Organization System (KOS) hierarchy and allows the user to explore the document sets and the KOSs at the same time. In this paper, we demonstrate the use of the ICE-Map Visualization in combination with a simple automatic indexer to visualize the semantic overlap between a KOS and a set of documents.

In this paper we present a demo for efficient detecting of visitor’s attention in museum environment based on the application of intelligent complex event processing and semantic technologies. Semantics is used for the correlation of sensors’ data via modeling the interesting situation and the background knowledge used for annotation. Intelligent complex event processing enables the efficient real-time processing of sensor data and its logic-based nature supports a declarative definition of attention situations.

The OWLGrEd ontology editor allows graphical visualization and authoring of OWL 2.0 ontologies using a compact yet intuitive presentation that combines UML class diagram notation with textual Manchester syntax for expressions. We present an extension mechanism for OWLGrEd that allows adding custom information areas, rules and visual effects to the ontology presentation thus enabling domain specific OWL ontology visualizations. The usage of OWLGrEd and its extensions is demonstrated on ontology engineering examples involving custom annotation visualizations, advanced UML class dia-gram constructs and integrity constraints in semantic database schema design.

Exposing data about customizable products is a challenging issue, because of the number of features and options a customer can choose from, and the many constraints that exist between them. These constraints are not tractable without automatic reasoning. But the configuration process, which helps a customer to make her choice, one step at a time, is a traversal of a graph of partially defined products - that is, Linked Data. This natural yet fruitful abstraction for product customization results in a generic configuration API, in use at Renault, who has begun publishing data about its range in this way. Current achievements and prototypes of forthcoming developments are presented.

The Linked Data cloud contains large amounts of RDF data generated from databases.

Citizens are increasingly aware of the influence of environmental and meteorological conditions on the quality of their life. This results in an increasing demand for personalized environmental information, i.e., information that is tailored to citizens’ specific context and background. In this demonstration, we present an environmental information system that addresses this demand in its full complexity in the context of the PESCaDO EU project. Specifically, we will show a system that supports submission of user generated queries related to environmental conditions. From the technical point of view, the system is tuned to discover reliable data in the web and to process these data in order to convert them into knowledge, which is stored in a dedicated repository. At run time, this information is transferred into an ontology-based knowledge base, from which then information relevant to the specific user is deduced and communicated in the language of their preference.

Detection rules represent one of the components of the rule models in event processing systems. These rules can be discovered from data using data mining techniques or domain experts’ knowledge. We demonstrate a system that provides its users the means for creating and validating such rules. The system is applied on real-life environmental scenarios, where the main source of data comes from sensors. Based on historical data about events of interest, the scope is to formulate rules that could have caused these events. Using a scalable infrastructure the rules can be tested on massive amount of data in order to observe how past events would fit to these rules. In addition, we create semantic annotations of the dataset and use them in the system outputs in order to support interoperability with other systems.

The Linked Open Data University of Münster (LODUM) project establishes a university-wide infrastructure to publish university data as Linked Open Data. The main goals are to increase visibility and accessibility of data produced and collected at the university, and to facilitate effective reuse of these data. This includes the goal to ease the development of applications and mashups based on the data, so that the common user can benefit from the LODUM data. This demo shows the LODUM infrastructure that facilitates application development, and two applications that demonstrate the potential of the LODUM data API.

Representing part-whole and mereotopological relations in an ontology is a well-known challenge. We have structured 23 types of part-whole relations and hidden the complexities of the underlying mereotopological theory behind a user-friendly tool:

OntoPartS

. It automates modelling guidelines using, mainly, the categories from DOLCE so as to take shortcuts in the selection process, and it includes examples and verbalizations to increase understandability. The modeller’s domain ontology, represented in any of the OWL species, can be updated automatically with the selected relation with a simple one-click button.

The AERS datasets is one of the few remaining, large publicly available medical data sets that until now have not been published as Linked Data. It is uniquely positioned amidst other medical datasets. This paper describes the Hubble prototype system for clinical decision support that demonstrates the speed, ease and flexibility of producing and using a Linked Data version of the AERS dataset for clinical practice and research.

A scientific conference is a type of event for which the structured program is generally known in advance. The Semantic Web community has setup a so-called Semantic Web dog food server that exposes structured data about the detailed program of more and more conferences and their sub-events (e.g. sessions). Conferences are also events that trigger a tremendous activity on social media. Participants tweet or post longer status messages, engage in discussion with comments, share slides and other media captured during the conference. This information is spread over multiple platforms forcing the user to monitor many different channels at the same time to fully benefit of the event. In this paper, we present Confomaton, a semantic web application that aggregates and reconciles information such as tweets, slides, photos and videos shared on social media that could potentially be attached to a scientific conference.

In this paper, we introduce OBA – an application for NLP-based annotation of natural language texts with ontology classes and relations. OBA provides support for different tasks required for semi-automatic semantic annotation. Among other things, it supports creating manual semantic annotations in order to enrich the set of lexical patterns, automatically annotating large corpora based on specified lexical patterns, and evaluating the results of semantic annotation.

An increasing amount of data represented using Resource Description Framework (RDF) have appeared on the Semantic Web.

We present DEFENDER and illustrate the benefits of identifying promising query decompositions and efficient plans that combine results from federations of SPARQL endpoints. DEFENDER is a query decomposer that implements a two-fold approach. First, triple patterns in a SPARQL query are decomposed into simple sub-queries that can be completely executed on one endpoint. Second, sub-queries are combined into a feasible bushy tree plan where the number of joins is maximized and the height of tree is minimized. We demonstrate DEFENDER and compare its performance with respect to state-of-the-art RDF engines for queries of diverse complexity, networks with different delays, and dataset differently distributed among a variety of endpoints.