Decision Support Tools for Ontological Engineering - Efficient Decision Support Systems: Practice and Challenges from Current to Future - page 64

4. Discussions
Applications of the two tools have been conducted and presented in (Guo & Chan, 2010)
and (Obst, 2009). Some strengths and weaknesses of the tools have been observed from the
application experiences.
Onto3DViz is able to successfully render the complex domain concepts and relationships
among concepts in a 3D model, thereby creating a visualized knowledge model that IMT
formulated. The user can easily examine and manipulate the 3D model by performing the
operations of zooming, rotating, and translating using the computer mouse or keyboard.
The 3D model clearly shows the hierarchical structure of the static knowledge; it can also
support representing the tasks associated with each object and arranging the tasks in the
correct order. However, some weaknesses were also observed in the current version of
Onto3DViz. First, when the visual objects are too close to each other, the objects and labels
become overlapped, which makes it difficult for users to clearly see details of the model.
Secondly, Onto3DViz does not currently support a search function. If the user wants to
retrieve a particular concept, he or she has to manually examine the 3D model so as to find
the corresponding visual object. This can be difficult if the ontology is complex or consists of
a large number of concepts. Thirdly, if an application ontology consists of too many
hierarchical levels, the lower level nodes will be small and difficult to see. This happens
because the scaling technique implemented in Onto3DViz (described in section 3.1.4)
reduces the size of the visual nodes in the lower levels.
The biggest limitation of the DFKE is the lack of integration with an ontology editor, such as
Protégé and Dyna. This limitation means the user would need to conduct ontology
modeling and ontology management as two different tasks and on two separate software
tools. This is not convenient and DFKE in its current version does not adequately support
the ontology authoring process.
5. Conclusions and future works
The objective of this research is to develop a suite of ontological engineering tools which
supports (1) static and dynamic knowledge visualization, and (2) management of an
application ontology. The motivation for this objective is derived from an analysis of
existing tools which reveals their lack of support for modeling and visualizing dynamic
knowledge. Among the few tools that support visualization, their capabilities for visualizing
a large volume of information is limited due to the constraints of the 2D graphical space.
Another weakness observed from our analysis is that existing ontology management
frameworks provide inadequate support for replication and evolution of ontology, and they
do not support detecting when a public domain ontology has been possibly tampered with.
To address these limitations, Onto3DViz has been developed to support 3D visualization of
an ontology model and the DFKE was designed for ontology management.
However, Onto3DViz and the monotonic DFKE have some weaknesses which will be
tackled in the future. The current version of Onto3DViz can be enhanced by adding a
collision avoidance system to address the problem of overlapping concepts. The collision
avoidance system would verify that the visual objects in a 3D model do not collide. And if a
collision is detected, it will automatically reassign the location of the visual objects and
adjust the space between them. To address the difficulty of finding concepts in a model, a
feature that supports automatic identification of a required concept can be added to
Onto3DViz. To prevent overly tiny visual nodes from being generated, the next version of