Abstract. This paper presents a context-driven approach that aims at supporting semantic mediation between composed Web services. Despite the widespread adoption of Web services by the IT community, innovative solutions are needed in order to overcome the challenging issue that relates to the semantic disparity of exchanged data. Indeed, there is a lack of means for interpreting these data according to the contextual requirements of each Web service. The context-driven approach suggests two steps. The first step consists of developing a model for anchoring context to data flowing between Web services. In the second step, we use this model to support the semantic mediation between Web services engaged in a composition. 1

Abstract. Semantic Web Services (SWS) aim at the automated discovery and orchestration of Web services on the basis of comprehensive, machineinterpretable semantic descriptions. In that, SWS strive for automated interoperability and reusability of heterogeneous services through matchmaking of semantic capability and interface descriptions. However, to do so, established SWS reference models build on the general assumption that either (a) SWS providers subscribe to a common vocabulary to annotate their services or (b) alignments between distinct vocabularies are established. This is due to the fact that SWS descriptions are lacking sufficient meaningfulness to automatically infer relationships between syntactically different semantic annotations. In order to address these issues and to overcome the need for (a) and (b), we propose a representational approach which allows to enrich standard SWS descriptions through vector spaces, which are represented as a dedicated ontology being aligned with existing SWS standards. As a result, similarities between instances used to annotate SWS become automatically computable by means of spatial distances. Hence, our approach significantly contributes to solve the interoperability problem between heterogeneous SWS as well as SWS reference models.

Abstract—Semantic Web Services (SWS) aim at the automated discovery and orchestration of Web services on the basis of comprehensive, machine-interpretable semantic descriptions. However, heterogeneities between distinct SWS representations pose strong limitations w.r.t. interoperability and reusability. Hence, semantic level mediation, i.e. mediation between concurrent semantic representations, is a key requirement to allow SWS matchmaking algorithms to compare capabilities of distinct SWS. In that, semantic level mediation requires to identify similarities across distinct SWS representations. Since current approaches to mediate between distinct service annotations rely either on manual one-to-one mappings or on semi-automatic mappings based on the exploitation of linguistic or structural similarities, these are perceived to be costly and error-prone. We propose a mediation approach enabling the implicit representation of similarities across distinct SWS by grounding these in socalled Mediation Spaces (MS). Given a set of SWS and their respective MS grounding, a general-purpose mediator automatically computes similarities to identify the most appropriate SWS for a given request. A prototypical application illustrates our approach.

Abstract. Composition of Web services is a cornerstone step in the development of interoperable systems. However, Web services still face data-heterogeneity challenges, although several attempts of using semantics. In addition, the context in which Web services evolve is still somehow ”ignored”, hampering their adaptability to changes in composition situations. In this paper, we argue how context permits to determine the semantics of interfaces that Web services expose to third parties. We show the need for a context- and semantic-based approach for Web services composition.

Keywords: Web service, data quality, data interpretation, semantics, context The Internet era has evolved from a Web of documents to a Web of services. Web services are intended to be application components that can be reused and integrated to create more advanced, innovative Web applications without needing to develop them from scratch. Unfortunately, Web services distributed on the Internet are usually independently developed by different organizations and/or individuals and have diverse assumptions about the interpretation of the exchanged data, such as inconsistent data representation and conceptualization. Such data misinterpretation can result in Data Quality (DQ) problems and hamper the potential of Web services. We identify important DQ challenges in Web services composition and present a classification of the resulting DQ problems. We suggest a novel reconciliation framework for addressing these problems and evaluate the framework in terms of scalability, adaptability, and extensibility. Finally, we identify important future directions in data quality and Web services. 1.

Bruce Spencer and Sandy Liu are the leader and a researcher in the Internet Logic group, which specializes in reasoning systems for Internet applications. The group’s activities on the Semantic Web include (i) a Semantic Web Lab with close ties to RuleML.org, (ii) a graduate course at the University of New Brunswick on the Semantic Web Techniques, (iii) BASeWEB (Business Agents and the Semantic Web) workshop held annually in conjunction with the Canadian AI Conference since 2002, (iv) an open source reasoning engine jDREW [2] on SourceForge, (v) a queuing inference engine [3], (vi) DeFleX, an XML Router for agile knowledge workflows [1], and (vii) WSIRD, a rule-based data integration engine between Web Services [4]. Challenges: Web Services can be seen as functional components that can be selected for composition to achieve certain purposes, based on descriptions of the components ’ purposes, their preconditions and effects, and the data they accept and produce. The selection may be done by an abstract reasoner, such as a planner, which uses these descriptions to infer that a specific combination of

Abstract. The rise of the Internet facilitates an ever increasing growth of virtual, i.e. digital spaces which co-exist with the physical environment, i.e. the physical space. In that, the question arises, how physical and digital space can interact synchronously. While sensors provide a means to continuously observe the physical space, several issues arise with respect to mapping sensor data streams to digital spaces, for instance, structured linked data, formally represented through symbolic Semantic Web (SW) standards such as OWL or RDF. The challenge is to bridge between symbolic knowledge representations and the measured data collected by sensors. In particular, one needs to map a given set of arbitrary sensor data to a particular set of symbolic knowledge representations, e.g. ontology instances. This task is particularly challenging due to the vast variety of possible sensor measurements. Conceptual Spaces (CS) provide a means to represent knowledge in geometrical vector spaces in order to enable computation of similarities between knowledge entities by means of distance metrics. We propose an approach which allows to refine symbolic concepts as CS and to ground ontology instances to so-called prototypical members which are vectors in the CS. By computing similarities in terms of spatial distances between a given set of sensor measurements and a finite set of CS members, the most similar instance can be identified. In that, we provide a means to bridge between the physical space, as observed by sensors, and the digital space made up of symbolic representations.

Abstract. Semantic Web Services (SWS) aim at the automated discovery and orchestration of Web services on the basis of comprehensive, machine-interpretable semantic descriptions. Since SWS annotations usually are created by distinct SWS providers, semantic-level mediation, i.e. mediation between concurrent semantic representations, is a key requirement for SWS discovery. Since semantic-level mediation aims at enabling interoperability across heterogeneous semantic representations, it can be perceived as a particular instantiation of the ontology mapping problem. While recent SWS matchmakers usually rely on manual alignments or subscription to a common ontology, we propose a two-fold SWS matchmaking approach, consisting of (a) a general-purpose semantic-level mediator and (b) comparison and matchmaking of SWS capabilities. Our semantic-level mediation approach enables the implicit representation of similarities across distinct SWS by grounding service descriptions in so-called Mediation Spaces (MS). Given a set of SWS and their respective grounding, a SWS matchmaker automatically computes instance similarities across distinct SWS ontologies and matches the request to the most suitable SWS. A prototypical application illustrates our approach.

Abstract. The increasing availability of sensor data through a variety of sensor-driven devices raises the need to exploit the data observed by sensors with the help of formally specified knowledge representations, such as the ones provided by the Semantic Web. In order to facilitate such a Semantic Sensor Web, the challenge is to bridge between symbolic knowledge representations and the measured data collected by sensors. In particular, one needs to map a given set of arbitrary sensor data to a particular set of symbolic knowledge representations, e.g. ontology instances. This task is particularly challenging due to the potential infinite variety of possible sensor measurements. Conceptual Spaces (CS) provide a means to represent knowledge in geometrical vector spaces in order to enable computation of similarities between knowledge entities by means of distance metrics. We propose an ontology for CS which allows to refine symbolic concepts as CS and to ground instances to so-called prototypical members described by vectors. By computing similarities in terms of spatial distances between a given set of sensor measurements and a finite set of prototypical members, the most similar instance can be identified. In that, we provide a means to bridge between the real-world as observed by sensors and symbolic representations. We also propose an initial implementation utilizing our approach for measurement-based Semantic Web Service discovery.

Service Oriented Computing (SOC) is a popular computing paradigm for the development of distributed Web applications. Service composition, a key element of SOC, is severely hampered by various types of semantic heterogeneity among the services. In this paper, we address the various semantic differences from the context perspective and use a lightweight ontology to describe the concepts and their specializations. Atomic conversions between the contexts are implemented using XPath functions and external services. The correspondences between the syntactic service descriptions and the semantic concepts are established using a flexible, standard-compliant mechanism. Given the naive BPEL composition ignoring semantic differences, our reconciliation approach can automatically determine and reconcile the semantic differences. The mediated BPEL composition incorporates necessary conversions to convert the data exchanged between different services. Our solution has the desirable properties (e.g., adaptability, extensibility and scalability) and can significantly alleviate the reconciliation efforts for Web services composition.