Information Technology Reference
In-Depth Information
2.1.4 Lack of support for ontology visualization
Most ontology editing tools lack visualization support. Often ontology visualization tools
use 2 dimensional on 2D graphics for representing concepts and relationships among
concepts. 2D graphics are often employed for representing static knowledge in hierarchical
or frame like structures. However, 2D graphics are not adequate for representing complex
and related information because there is insufficient space on a bi-dimensional plane, where
complex and related visual objects tend to squeeze together or even overlap with each other.
In an effort to accommodate a model of complex related objects in a 2D plane, existing
visualization tools either collapse lower priority visual objects into high level objects, or
implement different representations that offer multi-perspectives on an organization of
knowledge elements. While these approaches solve the problem of related objects being
overcrowded in a shared space, it sacrifices clarity of the display.
A popular ontology editor is Protégé, which has some degree of visualization capabilities in
that it generates a tree view of classes. The Protégé's ontology visualization applications are
implemented in its plug-ins; some examples of these ontology visualization plug-ins include
Jambalaya (Wu & Storey, 2000) and OntoSphere (Bosca et al., 2005), which are discussed as
follows.
Jambalaya is an ontology visualization tool plug-in in Protégé. It provides several viewing
perspectives for the ontology model, thereby enhancing user browsing, exploring and
interacting with a 2D ontology visualization. Jambalaya only visualizes the static knowledge
of classes and instances of an application ontology, it does not support dynamic knowledge
visualization. Furthermore, since Jambalaya is based on 2D graphics, the space it supports is
insufficient for rendering complex knowledge. In its representation, text labels and symbols
tend to overlap when the domain ontology is represented as a hierarchy involving many
levels of concepts. This deficiency means it is difficult for users to view and understand the
concepts and the relationships among concepts when the domain ontology is complex.
OntoSphere is also a Protégé plug-in. By employing 3D graphics for visualization,
OntoSphere extends the volume of space available for visualizing overcrowded concepts.
The 3-dimentional or 3D view is natural for humans. And its main advantage is that it
allows users to manipulate the visualized knowledge elements of the application ontology
by means of the actions of zooming, rotating and translating. Through physical
manipulation of the concepts, the user can better understand a complex ontology. For this
purpose, OntoSphere provides four scenes so that the user can observe a visualized
application ontology from multiple perspectives.
However, the weakness of OntoSphere include (1) it was not developed based on any
ontological engineering methodology, and (2) it does not support visualization of dynamic
knowledge. Although the employment of 3D graphics enlarges the space available for
rendering images, the problem of overlapping concepts and labels still exists when the
application ontology is complex.
2.1.5 Lack of support for ontology management
Currently ontology development within the SW is not managed in a systematic manner.
Anyone can create, reuse, and/or extend concepts in a distributed and heterogeneous
environment, and different versions of an ontology can be created, which results in
backwards compatibility problems. A system for ontology management is needed in order
to document, track, and distribute ontologies in such an environment. Existing ontology
Search WWH ::
Custom Search