=Paper=
{{Paper
|id=Vol-2043/paper-11
|storemode=property
|title=A Pattern-based Ontology for the Internet of Things
|pdfUrl=https://ceur-ws.org/Vol-2043/paper-11.pdf
|volume=Vol-2043
|authors=Aldo Gangemi,Raffaele Lillo,Giorgia Lodi,Andrea Giovanni Nuzzolese,Valentina Presutti
|dblpUrl=https://dblp.org/rec/conf/semweb/GangemiLLNP17
}}
==A Pattern-based Ontology for the Internet of Things==
https://ceur-ws.org/Vol-2043/paper-11.pdf
A pattern-based ontology for the Internet of
Things
Aldo Gangemi1,2 , Raffaele Lillo3 , Giorgia Lodi2 ,
?
Andrea Giovanni Nuzzolese1, , and Valentina Presutti1
1
Semantic Technology Laboratory, ISTC-CNR, Rome, Italy
2
LIPN, Université Paris 13, Sorbone Cité, UMR CNRS, France
3
Digital Transformation Team, Italian Government, Rome, Italy
Abstract. The Internet of Things (IoT) is about inter-networking real
word objects in order to foster data exchange and communication among
things. In this work we present the IoT Application Profile (IoT-AP) on-
tology with a particular focus on the pattern-based design methodology
used for modelling the ontology.
Keywords: internet of things, ontology design patterns, ontology re-use
1 Introduction
In recent years a lot of research has been carried out for realising a novel
paradigm known as the Internet of Things (IoT). Basically, the IoT is about
inter-networking real word objects (i.e. things), such as vehicles, roads, build-
ings, etc. The inter-networking is realised by providing those things with sensors,
actuators, and network connectivity for enabling data production and exchange.
Nevertheless, the enhancement of real word objects with intelligent behaviours
is still challenging. In this paper we present an ontology, named the IoT Appli-
cation Profile (IoT-AP) ontology, for representing and modelling the knowledge
within the domain of the Internet of Things. The focus of the paper is mainly on
the Ontology Design Patterns re-used for modelling the ontologies and on the
design methodology. The ontology is part of a wider ontology network, which
has been designed in the context of a project founded by the Italian Govern-
ment aimed at providing a big data framework for dealing with the Open Data
coming from the Italian Public Administration (PA). The rest of the paper is
organised as follows. Section 2 presents the related work, Section 3 presents the
IoT-AP ontology, Section 4 presents a usage scenario. Finally, Section 5 presents
the conclusions and future work.
2 Related Work
The main considerable effort to offer a comprehensive ontology for the IoT is
the Semantic Sensor Network ontology4 [2] (SSN), which is currently a W3C
?
Corresponding author: andrea.nuzzolese@istc.cnr.it
4
https://www.w3.org/TR/vocab-ssn/
2 Aldo Gangemi et al.
candidate recommendation. SSN is built on top of the the SOSA ontology5 and
describes sensors and their observations, the involved procedures, the studied fea-
tures of interest, the samples used to do so, and the observed properties, as well
as actuators. More recently, the IoT-Lite ontology6 , a W3C member submission,
is aimed at instantiating and extending the SSN ontology by introducing three
distinct concepts belonging to the IoT domain, i.e. objects, systems or resources
and services. Our solutions is inspired by the IoT and SSN ontologies. However,
we cannot directly re-use those ontologies as (i) both are still unstable being not
W3C recommendations yet, (ii) we have a soft requirement that encourages us
to design self-contained ontologies in order to avoid dependencies with external
ontologies that might evolve separately, thus, introducing inconsistencies.
3 Ontology overview
In next sub-sections we present the design methodology used and the resulting
IoT-AP ontology.
3.1 Design Methodology
IoT-AP is designed by following best design practices and pattern-based ontology
engineering aimed at extensively re-using Ontology Design Patterns (ODPs) [3]
for modelling ontologies. The design methodology that we followed is based on an
extension [6] of the eXtreme Design [1], an agile design methodology developed in
the context of the NeON project7 . Such an extension mainly focuses on providing
ontology engineer with clear strategies for ontology re-use. According to the
guidelines provided by [6], we adopted the indirect re-use. This means that ODPs
are used as templates. At the same time, the ontology guarantees interoperability
by keeping the appropriate alignments with the external ODPs, and provides
extensions that satisfy more specific requirements.
3.2 IoT-AP ontology
Table 1 reports the competency questions used at design time for modelling the
IoT-AP ontology. Figure 1 shows the core classes and properties of the IoT-AP
ontology by using the Graffoo notation.
The IoT-AP ontology is designed around the class Observation, which is
declared as subclass of dul:Event and is modelled by re-using the time indexed
situation8 Ontology Design Pattern (ODP) and the Observation9 ODP. In fact,
an observation represents the situation carried out by a sensor of estimating or
5
https://www.w3.org/2015/spatial/wiki/SOSA_Ontology
6
https://www.w3.org/Submission/2015/SUBM-iot-lite-20151126/
7
http://www.neon-project.org/nw/.
8
http://ontologydesignpatterns.org/wiki/Submissions:TimeIndexedSituation
9
http://ontologydesignpatterns.org/wiki/Submissions:Observation
A pattern-based ontology for the Internet of Things 3
Table 1. Competency questions used for modelling the ontology.
ID Competency question
CQ1 What is the sensor that makes an observation?
CQ2 What is the object (i.e. the feature of interest) associated with an observation?
CQ3 When is a certain observation made?
CQ4 What is the value of an observation?
CQ5 What is the observation parameter of an observation?
CQ6 What is the quality of an observation?
Fig. 1. Core classes of the IoT-Ont.
calculating a value of an attribute of a feature of interest at a specific time. Addi-
tionally, an observation can be associated with an instance of the class Geometry,
which identifies a location as a point, line, polygon, etc., and is expressed by
using coordinates in some coordinate reference system. This class Sensor is de-
clared as subclass of dul:Object and represents a sensor. A sensor is a device
whose purpose is to detect and respond to events or changes in its environment.
The detected information can be used and disseminated for successive elabora-
tions. Individuals of the class Sensor are associated with individuals of the class
Observation by means of the object property makesObservation. An observa-
tion can be associated with an ObservationParameter by means of the object
property hasObservationParameter. The class ObservationParameter repre-
4 Aldo Gangemi et al.
sents a property or a characteristic of a feature of interest under observation.
For instance, if we say that the temperature of the kitchen is 28 degrees cel-
sius, the observed parameter is the temperature. Additionally, an Observation
can be associated with a MeasurementQuality. The class MeasurementQuality
is declared as subclass of dul:Quality and expresses quality parameters or
measurement capabilities of an observation. Examples include latency, accuracy,
repeatability, etc. We re-used the Region ODP10 to represent and reason on val-
ues of attributes of MeasuramenteQuality and Observation. Hence, the object
property hasObservationValue links individuals of the class Observation to
individuals of the class ObservationValue and the object property hasMeasure-
mentQualityValue links individuals of the class MeasurementQualityValue.
The classes ObservationValue and MeasurementQualityValue represent val-
ues for observations and measurement quality, respectively. Additionally, both
classes are declared as subclasses of Value, which, in turn, is subclass of dul:Re-
gion. Collections of observations made by a sensor can be represented by in-
stantiating the class ObservationCollection. The conceptualisation around
ObservationCollection is obtained by re-using the Collection11 ODP. Hence,
ObservationCollection is subclass of dul:Collection and is associated with
the class Observation by means of the object property consistOf. The IoT-AP
ontology is aligned with DOLCE Ultra-Light and the Semantic Sensor Network
(SSN) ontology. The alignments with these two ontologies are kept in a separate
OWL file. Both the core IoT-AP ontology and the OWL file with the alignments
are available online12 . The HTML documentation is available via LODE13 .
4 Usage scenario
The following RDF triples, serialised as TURTLE, provide a usage example of
the IoT-AP ontology. These triples come from real traffic data provided us by
an organisation part of the Italian Public Administration.
:sensor a iotap:Sensor;
iotap:makesObservation :observation .
:observation a iotap:Observation;
iotap:hasFeatureOfInterest :a1_Calenzano-SestoFiorentino;
iotap:atTime :observation_time;
iotap:hasObservationParameter :average_speed;
iotap:hasObservatiobValue :average_speed_value .
:a1_Calenzano-SestoFiorentino a iotap:FeatureOfInterest;
rdfs:label "A1 road segment between Calenzano and Sesto Fiorentino".
:observation_time a iotap:TimeInterval;
iotap:startTime "2017-07-27T23:12:00+02:00"^^xsd:dateTime;
iotap:endTime "2017-07-28T01:54:00+02:00"^^xsd:dateTime .
10
http://ontologydesignpatterns.org/wiki/Submissions:Region
11
http://ontologydesignpatterns.org/wiki/Submissions:Collection
12
IoT-AP: http://stlab.istc.cnr.it/IoT-AP/IoT-AP.rdf - alignments: http://
stlab.istc.cnr.it/IoT-AP/IoT-AP-aligns.rdf
13
https://goo.gl/MFggc4
REFERENCES 5
:average_speed a iotap:ObservationParameter;
rdfs:label "Average speed" .
:average_speed_value a iotap:ObservationValue;
value "65";
iotap:hasMeasuramentUnit :kilometres_per_hour .
:kilometres_per_hour a iotap:MeasurementUnit;
rdfs:label "Kilometres per hour" .
The individual :sensor represents the sensor who makes the observation
identified by the individual :observation. This observation is about the average
speed observed by the sensor during the time interval identified by :obser-
vation time on the road segment of the A1 motorway between the cities of
Calenzano and Sesto Fiorentino (i.e. the feature of interest identified by the
individual :a1 Calenzano-SestoFiorentino). The value associated with the
observation (i.e. the individua :average speed value) allows to say that the
observed average speed is 65 Km/h.
5 Conclusion and future work
This paper proposes an ontology for the Internet of Things, namely the IoT-AP
ontology. The focus of the paper is on the pattern-based design methodology used
for modelling such an ontology. The design of the ontology is part of a project
with the Italian Government aimed at implementing a big data framework for
dealing with a variety of data sources including Open Data. As future work we
are planning to assess the quality of the ontology by using real data coming from
the big data framework as Linked Open Data and to analyse the ontology in the
context of the wider ontology network we are designing.
References
[1] Eva Blomqvist et al. “Experimenting with eXtreme Design”. In: Proc. of EKAW
2010. (Lisbon, Portugal). Springer, 2010, pp. 120–134.
[2] Michael Compton et al. “The SSN ontology of the W3C semantic sensor network
incubator group”. In: Web semantics: science, services and agents on the World
Wide Web 17 (2012), pp. 25–32.
[3] Aldo Gangemi and Valentina Presutti. “Ontology Design Patterns”. In: Handbook
on Ontologies, 2nd Ed. Springer, 2009, pp. 221–243.
[4] Aldo Gangemi et al. “Sweetening ontologies with DOLCE”. In: Knowledge engi-
neering and knowledge management: Ontologies and the semantic Web. Vol. 2473.
Lecture Notes in Computer Science. Berlin, Germany: Springer, 2002, pp. 166–
181. doi: 10.1007/3-540-45810-7_18.
[5] Sara Hachem, Thiago Teixeira, and Valérie Issarny. “Ontologies for the internet of
things”. In: Proceedings of the 8th Middleware Doctoral Symposium. ACM. 2011,
p. 3.
[6] Valentina Presutti et al. “The role of Ontology Design Patterns in Linked Data
projects”. In: Proc. of ER 2016. Springer, 2016, pp. 113–121.