Please use this identifier to cite or link to this item:
Title: Enhancing ontology quality through a pipelined extension of quality assessment and visualisation frameworks
Authors: Mc Gurk, Silvio
Keywords: Semantic Web
Ontologies (Information retrieval)
Issue Date: 2018
Citation: Mc Gurk, S. (2018). Enhancing ontology quality through a pipelined extension of quality assessment and visualisation frameworks (Master's dissertation).
Abstract: Ontologies have become increasingly important on the Semantic Web as a means of sharing concepts and data. The success of systems making use of ontology schemas depends mainly on the quality of their underlying ontologies. Assessing ontologies for their quality is thus crucial to determine quality problems and as a result improve upon them. Various researchers discuss how detection and correction of quality issues at an early stage is necessary to reduce costs and maintenance at later stages, which is more difficult to achieve and requires more effort. In this regard, a number of metrics have been proposed by researchers to measure different aspects of quality, accompanied by various frameworks and tools. This thesis aims at improving the quality assessment and evaluation of ontologies, proposing a system to identify poor quality aspects, and provide possible suggestions to improve the assessed ontology. A systematic survey on quality metrics applicable for ontologies in the Semantic Web domain is carried out. Existing quality assessment and visualisation frameworks are extended according to the results of the initial study. OntoQAV, an innovative pipeline linking together a quality assessment framework (Luzzu) and an ontology visualisation framework (WebVOWL) is implemented in order to establish an ecosystem whereby knowledge engineers can assess and interactively understand quality problems within concepts and properties in ontologies. In evaluating the pipeline, ontologies and Linked Data vocabularies are assessed and their quality investigated through the quality visualisation rendered by the pipeline. Usefulness and usability evaluations are also carried out to determine the extent to which the pipelined system is useful to stakeholders to assess the quality of ontologies and datasets to determine those which are fit-for-use, and to identify the degree at which the tool’s interface is user-friendly, its learnability and ease of use, respectively. Feedback provided by the evaluators indicate the need for such a tool, as an intuitive way of visualisation quality problems.
Appears in Collections:Dissertations - FacICT - 2018
Dissertations - FacICTAI - 2018

Files in This Item:
File Description SizeFormat 
  Restricted Access
3.63 MBAdobe PDFView/Open Request a copy

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.