Because the volume of web resources continues to increase exponentially, there
is a recognized need for advanced strategies that will enable resources to be
interpreted more logically by search engines and other intelligent agents (Chen
et al., 2008). The navigation through hundreds of the documents in
order to find the interesting information is a tough job and waste of the time
and effort (Binwahlan et al., 2009). As such,
there are many individuals and organizations beginning to explore the integration
of semantic web strategies into their future web development activities (Zhu
et al., 2009). Enabling reuse and sharing of creative knowledge in
engineering domain becomes our driving force to develop the knowledge map navigation.
Map-based navigation seems quite natural to humans because using a map is a
very convenient way to describe an environment and to share it with other people
(Yanqing et al., 2010). Knowledge visualization
and retrieval are the key techniques to implement effective knowledge management
(Chung et al., 2010).
A knowledge map is generally a visual representation of knowledge about
knowledge rather than of knowledge itself. It basically helps to detect
the sources of knowledge and structure of knowledge by representing the elements
and structural links of application domains. Some kind of knowledge map structure
can be found in literature are: hierarchical/radial knowledge map, networked
knowledge map, knowledge source map and knowledge flow map. Li
and Lu (2010) organized knowledge resources based on Intelligent Topic Map
(ITM) which extends the conventional topic map in structure and enhances its
reasoning functions. ITM establishes a novel multi-resource knowledge organization
model which depicts the hierarchical relationship of cluster-topic-knowledge
The challenge for the visual representation of the knowledge maps is to develop a visual tool for both navigating a large information space as well as for discovering possible contexts and relationships between groups of items. In our model, there are two major ways to enable the sharing of knowledge between users (Fig. 1).
As a matter of fact, some research effort has been attempted to employ ontology-based
techniques in manufacturing or in the production domain for integrating engineering
applications. Wei et al. (2009) developed an
Ontology-based Manufacturing Design System (OMDS) to present the PKM (Product
Knowledge model), realizing semantic match and enhancing the performance of
organization capability and knowledge sharing.
|| Research structure
Sun and Ding (2010) proposed the Feature Command Ontology
Model (FCOM) which includes a number of Neutral Semantic Feature Commands (NSFCs)
corresponding to the basic modeling operations of CAD systems. The FCOM which
has explicit ontological semantics, provides the uniform representation of heterogeneous
information and helps to shield the heterogeneity of data sources.
In the problem of design and manufacturing, functional knowledge represents designers intention. It plays a crucial role in engineering tasks such as designing and trouble shooting by engineers as well as understanding artifacts by users. In this study, we proposed functional knowledge ontology which can enable designers to better understand the design requirements and carry out effective reasoning in the functional space. That is, our goal firstly is to establish a functional ontology. The functional ontology enables us to realize a functional understanding system which identifies functional structures of an artifact.
In addition, concept maps are collections of propositions which can be seen
as simplified natural language sentences, displayed as a two-dimensional network
of labeled nodes and links (Erduran-Avci et al.,
2009). In this study, we propose a new approach that combines function model
and concept map to establish an ontology-based model for functional knowledge.
Further, the annual Taiwan Robot Game Contest started from 1997 is to cultivate
the interests among university and/or college students for creation and bring
out the potential in them for innovation. Competing robots would get awards
if the designs can demonstrate their outstanding mobility, agility and control
ability to overcome obstacles, such as avert light beams, simulated huge stones,
deep cliff and barriers, to reach out the destinations by four minutes. After
the annual game, all winners are requested to submit their abstract papers to
the RobotTW Portal (http://robotTW.ntust.edu.tw).
As for applied example of contest robot design knowledge, in order to share the functional knowledge model anywhere on the Internet, we use CmapTools which allow users to construct, navigate, share and criticize knowledge models represented as concept to implement that. It shows some benefits from functional ontology-based applications in engineering knowledge management domain.
The definition of ontology: Ontology means the subject of existence
in philosophy. In the context of knowledge sharing, ontology means an explicit
specification of a conceptualization (Maedche, 2001).
Ontology is a basic component for employing semantic web technology in knowledge
management systems. An ontology as a graph or network structure consists of
a set of concepts, relationships connecting concepts and instances assigned
to a particular concept. Sometimes classes and concepts
are used to represent the same in ontology. Ontology can be created using RDF
(Resource Description Framework) and OWL (Web Ontology Languages) recommended
by W3C consortium (Hayes et al., 2005).
In recent years, the ontology research about field knowledge gradually has been paid more attention in many ways. CIDOC National Digital Storage Program, Gene Ontology and seafood and grape wine ontology are the famous stories. However, engineering application of ontology has not yet been provoked, especially for knowledge sharing of design and manufacturing.
The functional ontology: Understanding an artifact is a major part of
engineering domain knowledge. Functionality is one of the key concepts in understanding
an artifact. While there is no common understanding of what a function is, people
share the idea would be tightly related to design intention. In contrast to
objective data about an artifact such as dimension, shape and structure, recognition
of functionality is dependent on systems, environments or situations in which
they are embedded. A function of a device explains what users can get using
it in an environment. A function of a component embedded in a system explains
why the component exists in the system and how it contributes to achieving the
systems whole-function. In the problem of design and manufacturing, functional
knowledge represents designers intention. It plays a crucial role in engineering
tasks such as designing and trouble shooting by engineers as well as understanding
artifacts by users (Kitamura et al., 2004).
Functional analysis is a means to provide the link between the systems architect
concept and the engineers design. It helps the engineers to develop a
complete design solution by providing a systematic means of identifying all
of the functional elements that need to be incorporated in the system. And then,
it provides an integrated view of the design to help the engineers understand
and manage the system under development as an integrated whole. Besides, the
purposes of the function model are to describe the functions and processes,
assist with discovery of information needs, help identify opportunities and
establish a basis for determining product and service costs (Kitamura
et al., 2006).
The development steps specifically for our functional ontology can be summarized
as follows (Noy and McGuiness, 2001):
||Determine the domain and scope of the ontology: The
domain and scope are restricted to the creative robot contest. The submitted
technical reports are used as the knowledge instances of the robot ontology.
We sketched a list of competency questions as follows to verify our knowledge
||What required parts make an integrated robot workable? What
is the core value of each part?
||For a specific robot team which mechanism is used to carry out a specific
||What kind of different mechanisms can fulfill fully the requirements of
movement and capture operations in the same time?
||If the target object changed to different shapes, a specific capture mechanism
designed for the past game theme still is in effect to complete its job?
||Is there unusual concept which fulfills some well-known actions?
||Enumerate important terms in the ontology: Currently
a list of all contest and robot terms is defined in Chinese and its English
names. There are really important to make problem-solving statements express
in a consistent way. It also provides the reference base to answer the following
questions: What properties do those terms have? What would we like to say
about those terms?
||Define the classes and the class hierarchy: The Classification
of robot ontology is developed with the combination method of top-down and
bottom-up approach. As stated in "Ontology Development 101", developing
the class hierarchy and defining properties of concepts (slots) are closely
intertwined. It is hard to do one of them first and then do the other. Therefore,
we tried to present few definitions of the major concepts in the hierarchy
first and then continue by describing their properties and so on. The functional
ontology of contest robot contains four main classes: Requirement,
Function, Structure and Triz Principle
||Define the properties of classes-slots: Slots are defined
by relationships of parts within and between other individual with intrinsic
properties or extrinsic properties. The facets of the slots are also needed
to define their slot cardinality and the slot-value types.
||Create instances: Instances can be created their relationships
based on slots (Fig. 3).
REPRESENTING ONTOLOGIES AS CMAPTOOLS
Concept maps are a intuitive visual knowledge representation technique. They
appear to have more "computational efficiency" than some other forms of knowledge
presentation such as pure text or predicate logic. Concept maps are graphs consisting
of nodes and arcs, representing relationships between nodes. The nodes are labeled
with descriptive text, representing the concept and the arcs are
labeled with the relationship type (Hayes and Eskridge, 2005).
|| An upper ontology of function for Contest Robot (generated
by Protégé software)
|| Instance implementation (generated by Protégé
Concept maps are informal representations that facilitate knowledge capture for human examination and sharing. CmapTools allows users to build knowledge models consisting of sets of interconnected concept maps annotated with material such as text documents, diagrams and video clips. It provides rich, searchable, browsable knowledge models available for navigation and collaboration across geographically-distant sites. It supports both synchronous and asynchronous methods for collaborative map development.
COE (CMapTools Ontology Editor) is comprised of a set of Java modules that
extend CmapTools with features that are particularly oriented towards the expression
of formalized ontologies while retaining and extending the human-oriented advantages
of the concept map interface (Novak and Canas, 2004).
|| Using COE to create Concept Map
|| Portion of a knowledge model developed by cmaptools for contest
The purpose is to give domain experts the ability to construct, share and examine
Web ontologies. The system supports importing, editing and storing of OWL Ontologies
in the form of concept maps, exporting concept maps to OWL/XML files, searching
for concepts and properties in existing concept maps and Web ontologies and
examining the different roles which concepts play in other ontologies which
gives composers the ability to choose appropriate concepts for their ontologies
(Leake et al., 2004).
To bridge the gap between the informal nature of concept maps and the formal, machine-readable Web ontology languages, COE uses a set of conventions and guidelines that enables users to construct syntactically valid Web ontologies using the concept-mapping interface. These conventions retain as far as possible an intuitive reading of the concept map while faithfully capturing the precision of the OWL syntax and are based on a few basic ideas. English words and phrases are used as far as possible and it avoids the mathematical logic terminology that pervades the OWL documentation. These conventions retain as far as possible an intuitive reading of the concept map while faithfully capturing the precision of the OWL syntax and are based on a few basic ideas. For example, the fact that motor are driving element would be represented in OWL-XML syntax by saying that Motor is a subclass of Driving Element; in COE, as shown in Fig. 4, it is rendered using a link labeled are from the subject to the object of the sentence, mirroring the syntax of a simple English sentence and avoiding the use of the subclass terminology.
Users can construct ontologies by direct manipulation of the CmapTools interface but for several of the OWL constructs, this requires a number of steps to make the appropriate linkages and set the box and line styles according to convention.
Figure 5 shows several opened windows, the result of navigating through our model based on functional perspective. By clicking the concept map icon underneath a concept, a list of available Cmaps is displayed which the user can select and open. It can be shared the functional knowledge model anywhere on the Internet.
In this study, we have developed a knowledge map navigation which is based on functional ontology and concept maps. A functional ontology model is presented to show up the relationships among diverse technical documents. Besides, present implementation of document model would be demonstrated by CmapTools which allow the users to export ontologies in OWL format.
Further, as for applied example of contest robot, it says some benefits from our proposed model in engineering knowledge management domain.
The authors would like to thank the National Science Council for its financial support.