Semantic IoT: The Key to Realizing IoT Value
97
integrate several RDF datasets, such as US census data, DBLP4, DBPedia3, FOAF5,
etc. Two major constraints in the design and development of semantic ontologies are
the quantity and the complexity that directly affects the maintainability, reusability,
understanding, integration, and applicability [37]. Ontology complexity, in general,
is the difficulty in designing, developing, modifying, and reusing the ontology which
can be measured using some software metric standard such as Weyuker’s criteria
[38]. There are other ontologies measuring matrices namely the number of the leaf
class, the number of root classes, and the average depth of inheritance tree used to
measure the ontology cohesiveness [39], entropy-based metric to measure ontology’s
structural complexity described as a UML diagram [40], etc. These metrics focus on
either one or two factors of ontology such as the structural complexity or empirical
validations. In this regard, a meta-ontology O2 matrix characterizing the semiotic
objects to identify three ontology measuring parameters such as structural, functional
and usability-profiling has been proposed [41]. Similarly, more than two measuring
metrics have been proposed by other researchers as well with some of them have a
limited utility [42, 43]. Nevertheless, these ontology metrics need to follow a few
guidelines such as name anonymous classes, name anonymous individuals, materialize the subsumption hierarchy and unify names, etc. They need to propagate
instances to the deepest possible class or property within the hierarchy and normalize
properties while creating the metrics. These analyses show the absence of systematic
methods and measuring parameters hence need to be addressed by future researchers.
6 Network Tools for Efficient SIoT
There are several semantic modeling tools have been emerged to facilitate the SIoT
domain. The ‘Hyper Thing’ is considered to be a semantic web URI validator to
distinguish between the URI identities a RealWorld Object and a web document
resource. It checks whether the method of publishing of URIs follows 303 URI
or the W3C hash practices. It also attempts to verify the redirection between the
Document URIs and the Real-World Object URIs to get rid of the data publisher
mistakenly redirect. The ‘NeON’ is a network tool that provides the methodology
for Ontology while the ‘OWL’ tool allows project for semantic validation with ontologies written in OWL/XML, RDF/XML, OWL Functional Syntax, OBO Syntax,
Manchester OWL Syntax, and KRSS Syntax. On the other hand, the ‘OQuaRE’
happens to be a square-based approach to evaluate an ontology quality based on
software quality evaluation and software quality requirement specifications. The
‘OntoClean’ tool in a problem domain defines the meta-properties for the construction of ontology language descriptions. Other similar tools find useful in the field of
SIoT are OnToology, Oops Ontocheck, OntoAPI, OntoMetric, Prefix, etc. Table 3
provides describes these tools in brief.
97
integrate several RDF datasets, such as US census data, DBLP4, DBPedia3, FOAF5,
etc. Two major constraints in the design and development of semantic ontologies are
the quantity and the complexity that directly affects the maintainability, reusability,
understanding, integration, and applicability [37]. Ontology complexity, in general,
is the difficulty in designing, developing, modifying, and reusing the ontology which
can be measured using some software metric standard such as Weyuker’s criteria
[38]. There are other ontologies measuring matrices namely the number of the leaf
class, the number of root classes, and the average depth of inheritance tree used to
measure the ontology cohesiveness [39], entropy-based metric to measure ontology’s
structural complexity described as a UML diagram [40], etc. These metrics focus on
either one or two factors of ontology such as the structural complexity or empirical
validations. In this regard, a meta-ontology O2 matrix characterizing the semiotic
objects to identify three ontology measuring parameters such as structural, functional
and usability-profiling has been proposed [41]. Similarly, more than two measuring
metrics have been proposed by other researchers as well with some of them have a
limited utility [42, 43]. Nevertheless, these ontology metrics need to follow a few
guidelines such as name anonymous classes, name anonymous individuals, materialize the subsumption hierarchy and unify names, etc. They need to propagate
instances to the deepest possible class or property within the hierarchy and normalize
properties while creating the metrics. These analyses show the absence of systematic
methods and measuring parameters hence need to be addressed by future researchers.
6 Network Tools for Efficient SIoT
There are several semantic modeling tools have been emerged to facilitate the SIoT
domain. The ‘Hyper Thing’ is considered to be a semantic web URI validator to
distinguish between the URI identities a RealWorld Object and a web document
resource. It checks whether the method of publishing of URIs follows 303 URI
or the W3C hash practices. It also attempts to verify the redirection between the
Document URIs and the Real-World Object URIs to get rid of the data publisher
mistakenly redirect. The ‘NeON’ is a network tool that provides the methodology
for Ontology while the ‘OWL’ tool allows project for semantic validation with ontologies written in OWL/XML, RDF/XML, OWL Functional Syntax, OBO Syntax,
Manchester OWL Syntax, and KRSS Syntax. On the other hand, the ‘OQuaRE’
happens to be a square-based approach to evaluate an ontology quality based on
software quality evaluation and software quality requirement specifications. The
‘OntoClean’ tool in a problem domain defines the meta-properties for the construction of ontology language descriptions. Other similar tools find useful in the field of
SIoT are OnToology, Oops Ontocheck, OntoAPI, OntoMetric, Prefix, etc. Table 3
provides describes these tools in brief.