Barry Norton

A compositional operational semantics for OWL-S

Software composition via workflow specifications has received a great deal of attention recently. One reason is the high degree of fit with the encapsulation of software modules in service-oriented fashion. In the Industry, existing workflow languages have been merged to form WS-BPEL, the Business Process Execution Language for Web Services. In the Research community OWL-S, a ontology for web services, has been submitted for standardisation alongside OWL, the Web Ontology Language in which it is expressed. The OWL-S Process Model is based on an abstraction of the common features of industrial workflow languages. On the one hand, WS-BPEL has only informal semantics; on the other, the type of semantics given to ontology-based work tends to be structural rather than computationally oriented. As a result the semantics developed for DAML-S, which led to OWL-S, are still deficient in some regards. In this paper we shall survey the existing semantics and introduce a novel semantics for the latest version of OWL-S that is focussed on the principle of compositionality, so far not tackled.

Combining human and computation intelligence: the case of data interlinking tools

Interlinking is without doubt one of the most active and mature areas of research and development in semantic technologies. Over the last decade or more a multitude of approaches to match, merge and integrate ontologies, both at the schema and instance levels have been proposed and successfully applied to resolve heterogeneity issues and, more recently, to interlink RDF data sets exposed over the Web as part of the Linked Open Data Cloud. The strengths and weaknesses of existing interlinking solutions, as well as their natural limitations and principled combinations have been intensively studied, not least through community projects such as the Ontology Alignment Evaluation Initiative. Human input remains a key ingredient of the process, either as a source of domain knowledge used to train matching algorithms and to build the underlying knowledge base, or to validate automatically computed results. In this paper we describe how such human input could be acquired and used to enhance the results of existing data interlinking technology via crowdsourcing. In a survey of data interlinking tools we identify several aspects of the interlinking process that crucially rely on human contributions and explain how these aspects could be subject to a semantically enabled human computation architecture that can be set-up by extending interlinking platforms such as Silk with direct interfaces to popular microtask platforms such as Amazons Mechanical Turk.

A Compositional Semantic Theory for Synchronous Component-Based Design

Digital signal processing and control (DSPC) tools allow application developers to assemble systems by connecting predefined components in signal-flow graphs and by hierarchically building new components via encapsulating sub-graphs. Run-time environments then dynamically schedule components for execution on some embedded processor, typically in a synchronous cycle-based fashion, and check whether one component jams another by producing outputs faster than can be consumed. This paper develops a process-algebraic model of coordination for synchronous component-based design, which directly lends itself to compositionally formalising the monolithic semantics of DSPC tools. By uniformly combining the well-known concepts of abstract clocks, maximal progress and clock-hiding, it is shown how the DSPC principles of dynamic synchronous scheduling, isochrony and encapsulation may be captured faithfully and compositionally in process algebra, and how observation equivalence may facilitate jam checks at compile-time.

Mapping Keywords to Linked Data Resources for Automatic Query Expansion

Linked Data is a gigantic, constantly growing and extremely valuable resource, but its usage is still heavily dependent on (i) the familiarity of end users with RDF’s graph data model and its query language, SPARQL, and (ii) knowledge about available datasets and their contents. Intelligent keyword search over Linked Data is currently being investigated as a means to overcome these barriers to entry in a number of different approaches, including semantic search engines and the automatic conversion of natural language questions into structured queries. Our work addresses the specific challenge of mapping keywords to Linked Data resources, and proposes a novel method for this task. By exploiting the graph structure within Linked Data we determine which properties between resources are useful to discover, or directly express, semantic similarity. We also propose a novel scoring function to rank results. Experiments on a publicly available dataset show a 17% improvement in Mean Reciprocal Rank over the state of the art.

Orchestration of semantic web services for large-scale document annotation

Armadillo is a tool that provides automatic annotation for the Semantic Web using unannotated resources like the existing Web for information harvesting, that is: combining a crawling mechanism with an extensible architecture for ontology population. The latter is achieved via largely unsupervised machine learning, boot-strapped from oracles, such as web-site wrappers. It is backed up by ‘evidential reasoning, which allows evidence to be gained from the redundancy in the Web as well as inaccuracies in information, also characteristic of todays Web, to be circumvented. In this paper we sketch how the architecture of Armadillo has now been reinterpreted as workflow templates that compose semantic web services and show how the porting of Armadillo to new domains, and furthermore the application of new tools, has thus been simplified and benefits from semantic discovery and automatic orchestration.

Reactive types for dataflow-oriented software architectures

Digital signal processing (DSP) tools, such as Ptolemy, LabView and iConnect, allow application developers to assemble reactive systems by connecting predefined components in generalised dataflow graphs and by hierarchically building new components by encapsulating subgraphs. We follow the literature in calling this approach dataflow-oriented development. Our previous work has shown how a new process calculus, uniting ideas from previous systems within a compositional theory, can be formally shown to capture the properties of such systems. This paper first recasts the graphical dataflow-oriented style of design into an underlying textual architecture design language (ADL) and then shows how the previous modelling approach can be seen as a system of process algebraic behavioural types for such a language, so that type checking is the mechanism used to statically diagnose the reactivity of applications. We show how both the existing notion of behavioural equivalence and a new behavioural preorder are involved in this judgement.

Scalable discovery of Linked APIs

A number of approaches bring together the principles and technologies which define Linked Data with those of RESTful services. Services and APIs are thus enriched by, and contribute to, the Web of Data. These approaches, referred to as Linked APIs, use graphic patterns as an intuitive way to describe input and output expectations. To enable agents to discover Linked APIs we propose metrics tailored for a scalable discovery system to measure the degree service descriptions match service templates formulated by an agent. The metrics are based on the used vocabularies in the service descriptions and templates as well as on the containment relation of the employed graph patterns. We introduce a cloud-based implementation of our system in a distributed environment to further address scalability. The results of our evaluation show the positive effects of a distributed computation strategy on the performance of our system.

Quality of service profiles in web service discovery

Standardization of the description and delivery of XML-based web services has opened up a market in ‘commercial off-the-shelf (COTS) software components. As a result, standardization efforts are being made towards the assembly of systems from web services where the coordination is defined by workflow languages. With several potential implementations for many of the tasks within such a system an automated discovery process is required. With many functional equivalents, it is necessary to discriminate between these on the basis of cost and performance.