20
A. Chatzimichail et al.
version of OWL Full. In this grammar classes can also be considered as instances
and they are generalized to refer to a broader subset of OWL vocabulary. Driven by
the entailments of the pD
∗ grammar and DLP, the OWL 2 RL profile of semantics is
realised as a partial axiomatisation of the OWL 2 semantics in the form of first-order,
known as OWL 2 RL/RDF rules. Rules defined by users over the ontology allow
richer semantic relationships to be articulated outside the descriptive capabilities of
OWL, combined with ontological awareness and rules.
SPARQL [128] is a declarative language which the W3C recommends to extract
and update information within RDF graphs. It is an expressive language, which
describes complex relations between entities. The syntax and difficulty of the
SPARQL query language have been studied relatively technically, showing that both
SPARQL algebra and relational algebra share the same expressive power [129].
SPARQL is mainly known as a query language for RDF, however it can define
SPARQL rules by using the CONSTRUCT graph format, which can generate new
RDF statements by merging existing RDF graphs. These rules are described in terms
of a CONSTRUCT and a WHERE clause: CONSTRUCT specifies the graph patterns, that is the set of RDF triple patterns that should be ingested to the underlying
RDF graph when the graphs in the WHERE clause fit successfully.
Finally, the SPARQL Inferencing Notation (SPIN) [130] is an attempt to simplify
the interpretation and execution of SPARQL rules on top of RDF graphs. Using
SPIN, SPARQL queries can be stored as RDF triples along with any RDF ontology,
allowing RDF instances to be connected to the related SPARQL queries, as well as
sharing and reuse of SPARQL queries. SPIN follows the interpretation of SPARQL
inference rules that can be used by iterative rules implementations to extract new
RDF statements from existing ones.
3.3.2 Fusion
One of the most important issues in the IoT sensor networks is the data management.
Sensor networks are facing resource constraints problems due to low battery power,
limited data processing capabilities, limited communication resources and a small
amount of memory. Furthermore, in many applications there are data coming from
many heterogeneous data sources that needs to be compared, combined and correlated
between each other. In this way, depending on the applications appropriate data
aggregation systems must be implemented for the processing of the data at the edge
or in the cloud.
Semantic Web helps enabling interoperability among data from different sources
through the content annotation. To become retrievable, data coming from sensor
network systems should be annotated.
The fusion through semantic technologies is realised through different structures.
For example, in text fusion, it is important to fuse statements and assertions from
different sentences, tables, or paragraphs to define definitions, objects, and their
semantic relationships. Another important role for semantic fusion is when there is
a need to fuse ontologies. The majority of researchers are using available ontologies
A. Chatzimichail et al.
version of OWL Full. In this grammar classes can also be considered as instances
and they are generalized to refer to a broader subset of OWL vocabulary. Driven by
the entailments of the pD
∗ grammar and DLP, the OWL 2 RL profile of semantics is
realised as a partial axiomatisation of the OWL 2 semantics in the form of first-order,
known as OWL 2 RL/RDF rules. Rules defined by users over the ontology allow
richer semantic relationships to be articulated outside the descriptive capabilities of
OWL, combined with ontological awareness and rules.
SPARQL [128] is a declarative language which the W3C recommends to extract
and update information within RDF graphs. It is an expressive language, which
describes complex relations between entities. The syntax and difficulty of the
SPARQL query language have been studied relatively technically, showing that both
SPARQL algebra and relational algebra share the same expressive power [129].
SPARQL is mainly known as a query language for RDF, however it can define
SPARQL rules by using the CONSTRUCT graph format, which can generate new
RDF statements by merging existing RDF graphs. These rules are described in terms
of a CONSTRUCT and a WHERE clause: CONSTRUCT specifies the graph patterns, that is the set of RDF triple patterns that should be ingested to the underlying
RDF graph when the graphs in the WHERE clause fit successfully.
Finally, the SPARQL Inferencing Notation (SPIN) [130] is an attempt to simplify
the interpretation and execution of SPARQL rules on top of RDF graphs. Using
SPIN, SPARQL queries can be stored as RDF triples along with any RDF ontology,
allowing RDF instances to be connected to the related SPARQL queries, as well as
sharing and reuse of SPARQL queries. SPIN follows the interpretation of SPARQL
inference rules that can be used by iterative rules implementations to extract new
RDF statements from existing ones.
3.3.2 Fusion
One of the most important issues in the IoT sensor networks is the data management.
Sensor networks are facing resource constraints problems due to low battery power,
limited data processing capabilities, limited communication resources and a small
amount of memory. Furthermore, in many applications there are data coming from
many heterogeneous data sources that needs to be compared, combined and correlated
between each other. In this way, depending on the applications appropriate data
aggregation systems must be implemented for the processing of the data at the edge
or in the cloud.
Semantic Web helps enabling interoperability among data from different sources
through the content annotation. To become retrievable, data coming from sensor
network systems should be annotated.
The fusion through semantic technologies is realised through different structures.
For example, in text fusion, it is important to fuse statements and assertions from
different sentences, tables, or paragraphs to define definitions, objects, and their
semantic relationships. Another important role for semantic fusion is when there is
a need to fuse ontologies. The majority of researchers are using available ontologies