Refine
Year of publication
- 2011 (3) (remove)
Document Type
- Bachelor Thesis (1)
- Doctoral Thesis (1)
- Study Thesis (1)
Keywords
- OWL <Informatik> (3) (remove)
Institute
Model-Driven Engineering (MDE) aims to raise the level of abstraction in software system specifications and increase automation in software development. Modelware technological spaces contain the languages and tools for MDE that software developers take into consideration to model systems and domains. Ontoware technological spaces contain ontology languages and technologies to design, query, and reason on knowledge. With the advent of the Semantic Web, ontologies are now being used within the field of software development, as well. In this thesis, bridging technologies are developed to combine two technological spaces in general. Transformation bridges translate models between spaces, mapping bridges relate different models between two spaces, and, integration bridges merge spaces to new all-embracing technological spaces. API bridges establish interoperability between the tools used in the space. In particular, this thesis focuses on the combination of modelware and ontoware technological spaces. Subsequent to a sound comparison of languages and tools in both spaces, the integration bridge is used to build a common technological space, which allows for the hybrid use of languages and the interoperable use of tools. The new space allows for language and domain engineering. Ontology-based software languages may be designed in the new space where syntax and formal semantics are defined with the support of ontology languages, and the correctness of language models is ensured by the use of ontology reasoning technologies. These languages represent a core means for exploiting expressive ontology reasoning in the software modeling domain, while remaining flexible enough to accommodate varying needs of software modelers. Application domains are conceptually described by languages that allow for defining domain instances and types within one domain model. Integrated ontology languages may provide formal semantics for domain-specific languages and ontology technologies allow for reasoning over types and instances in domain models. A scenario in which configurations for network device families are modeled illustrates the approaches discussed in this thesis. Furthermore, the implementation of all bridging technologies for the combination of technological spaces and all tools for ontology-based language engineering and use is illustrated.
In this study, the author evaluates four visualization methods (FlexViz, the Jambalaya applet, Experimental jOWL TouchGraph, and Plone ontology) from the pool of currently available web based visualization methods.
Further, this study attempts to answer of the following research questions: 1. What are the requirements for ontology visualization? 2. How is comparative analysis and evaluation carried out? 3. How can a chosen method be tested? Based on the results of the evaluation, the visualization method FlexViz was found to be the most suitable for the given scenario. The deployment of FlexViz was carried out and integrated within Plone CMS and logical parts of the VCD ontology are tested. Finally, FlexViz was analyzed from different stakeholder- perspectives. Results showed that FleViz is a reliable tool for visualizing, understanding and analyzing a developed ontology because of its user-friendly and interactive interface, but performs poorly in regards to the editing, extension and development process.
SPARQL can be employed to query RDF documents using RDF triples. OWL-DL ontologies are a subset of RDF and they are created by using specific OWL-DL expressions. Querying such ontologies using only RDF triples can be complicated and can produce a preventable source of error depending on each query.
SPARQL-DL Abstract Syntax (SPARQLAS) solves this problem using OWL Functional-Style Syntax or a syntax similar to the Manchester Syntax for setting up queries. SPARQLAS is a proper subset of SPARQL and uses only the essential constructs to obtain the desired results to queries on OWL-DL ontologies implying least possible effort in writing.
Due to the decrease in size of the query and having a familiar syntax the user is able to rely on, complex and nested queries on OWL-DL ontologies can be more easily realized. The Eclipse plugin EMFText is utilized for generating the specific SPARQLAS syntax. For further implementation of SPARQLAS, an ATL transformation to SPARQL is included as well. This transformation saves developing a program to directly process SPARQLAS queries and supports embedding SPARQLAS into running development environments.