=Paper=
{{Paper
|id=Vol-2228/short5
|storemode=property
|title=I2oTology - Tracking-Oriented Ontology
|pdfUrl=https://ceur-ws.org/Vol-2228/short5.pdf
|volume=Vol-2228
|authors=Levy Marlon Souza Santiago,Jauberth Weyll Abijaude,Péricles de Lima Sobreira,Fabíola Gonçalves Pereira Greve
|dblpUrl=https://dblp.org/rec/conf/ontobras/SantiagoASG18
}}
==I2oTology - Tracking-Oriented Ontology==
https://ceur-ws.org/Vol-2228/short5.pdf
I2oTology - Tracking-Oriented Ontology
Levy Marlon Souza Santiago1 , Jauberth Weyll Abijaude1 ,
Péricles de Lima Sobreira1 ,
Fabı́ola Gonçalves Pereira Greve1
1
Departamento de Ciências Exatas e Tecnológicas
Universidade Estadual de Santa Cruz (UESC) – Ilhéus, BA – Brazil
levyssantiago@gmail.com,
{jauberth,plsobreira}@uesc.br,
fabiola@ufba.br
Abstract. To join the Internet of Things (IoT) and Ontology concepts today it is
becoming a good strategy to save sensors and Smart Objects (SO) information
using all the semantic capabilities and ontology inferences to improve and give
some intelligence at the information manipulation, IoT-Lite and SSN (Semantic
Sensor Network) are examples of ontologies for IoT. This paper presents the
I2oTology, which is a tracking-oriented ontology. The I2oTology purpose is to
presents a semantic aimed at tracking smart objects based on some IoT-Lite
classes. It was made a simple test with this ontology but there is some classes,
properties and situations to be tested and also to know how much the ontology
is right, these topics will be considered at the future work.
1. Introduction
Internet of Things (IoT) adopts novel processing, communication architecture, smart tech-
nologies and management strategies to seamlessly integrate a large number of smart ob-
jects with the Internet [Li et al. 2018]. Service-oriented architectures (SOA) and method-
ologies have been widely adopted and studied in distributed systems, well before the
emergence of IoT. However, due to huge number of entities and large diverse service pool
in IoT, the trend has shifted towards using more lightweight services. Traditional SOAP-
based services have been gradually replaced by RESTful services and APIs are now the
new players in this field. APIs are easier to define, invoke, share, and monitor compared
to other service definition methods [Khodadadi and Sinnott 2017].
One of the most highlighting features of the Internet of Things domain is the het-
erogeneity of the information. One method to accomplish this interoperability is through
the usage of semantic-based technologies to annotate all the information shared by the
platforms [Agarwal et al. 2016]. The ontology alignment allows organizations to model
their own knowledge without having to stick to a specific standard [Gil et al. 2018].
Ontology is a representation vocabulary, often specialized to some domain or sub-
ject matter. More precisely, it is not the vocabulary as such that qualifies as an ontol-
ogy, but the conceptualizations that the terms in the vocabulary are intended to capture
[Chandrasekaran et al. 1999]. The main benefit of having an ontology for a specific do-
main is for confederacy and dissemination of knowledge about the domain and connecting
with other domains [Keat and Shahrir 2017]. The OWL (Web Ontology Language) is a
ontology language which facilitates greater machine interpretability of Web content than
that supported by XML, RDF, and RDF Schema (RDF-S) by providing additional vocab-
ulary along with a formal semantics [McGuinness et al. 2004].
The IoT-Lite is a lightweight semantic model for IoT proposed by
[Bermudez-Edo et al. 2016]. It’s an instantiation of the Semantic Sensor Network (SSN)
ontology [Compton et al. 2012]. The intent of IoT-Lite is not to be a full ontology for the
IoT, it was built to be a core lightweight ontology that allows relatively fast annotation
and processing time.
Hermes Widget IoT, [Veiga et al. 2017], is a component that extends the Hermes
context management system [Sene Júnior et al. 2014] representation layer allowing it to
handle information obtained from any sensor with a web endpoint, the Hermes Widget
IoT uses the semantics of IoT-oriented ontologies such as IoT-Lite and SSN, it allows
any context provider object, for example, a sensor, to be located, used, and have its cor-
responding context information represented and made available for querying through the
Internet, the geo:Point class [Brickley 2006] allows the system define a geographic loca-
tion of these context provider objects. The I2otology also integrates this class but is still
under tests, but the idea is to locate the SO’s as the Hermes Widget does.
A semantic Industrial Internet of Things (IIoT) architecture is proposed by
[Pease et al. 2017]. Between the architecture layers, there is an ontology called IIoT on-
tology which uses RDF and OWL for knowledge engineering. This ontology contains the
tracking ontology class IndoorTrackingDevices used to link device type to service, for
service functionality discovery. Also there is an inventory of assets which can be located
in real time, so the system “knows” when an asset is moving. The I2otology follows the
same concept, it can give some information to the system that is using it about which SO
is moving, who is moving it and where is its actual location. But the idea is to use RFID
portals in rooms, in this way if the Smart Object get out of a room, the system will “know”
that this object is moving because its RFID tag will be read.
The remainder of this paper is organized as follows. The next section is an ex-
planation about the I2oTology, the merging with IoT-Lite and a brief inference example.
The Section 3 describes the I2oTology implementation in a web system. Conclusions and
future work direction in Section 4.
2. The I2oTology
The I2oTology (Figure 1) is an extention of IoT-Lite. There’s some IoT-Lite classes that
wasn’t used, and there’s some classes added (Table 1) to give some others capabilities
which will be explained as follow. The proposed ontology presents a semantics aimed
at tracking smart objects making use of IoT-Lite location classes, as well as adding the
possibility of a reasoner to discover through this ontology if a specific object can be in a
given room with a certain person.
This ontology has been applied within a specific context (an university), but in the
question of tracking it can be said that it can be applied to similar contexts. The I2oTology
was tested (next section) with a web system which is still being developed to be used at
the UESC University.
Table 1. I2oTology’s classes explanations
This class represents institutions where this ontology is applied, in this case, the UESC
Institution
University.
This is a superclass that represents a general place, it can be a classroom, a laboratory and
Place
others.
Laboratory This class represents laboratory rooms and for this reason, it’s subclass of Place.
This one represents classrooms and as well as the Laboratory class, this is a subclass of
ClassRoom
Place.
This is an equivalent class1 . When an smart object individual “canBeIn” some place (e.g.
AllowedPlace Laboratory) and has the relation hasLocation to some kind of that place (e.g. individual
“lab 18”), so the reasoner will infer that this object is in the allowed place.
Person This class represents people. It’s used to define the person that is moving some object.
Teacher This class represents teachers and is subclass of Person.
Functionary A class that represents institution functionaries. Its also a subclass of Person.
Student Students representation, also a subclass of Person.
This class follows the same idea of AllowedPerson. It’s an equivalent class that represents
AllowedPerson if the object is been moved (using the canBeMovedBy and isMovedBy properties) by an
allowed person.
RfidSensor Represents Rfid readers. It is a subclass of SensingDevice
This class represents some rfid characteristics, for example, it’s Antenna(s) and this
RfidSensorCharacteristic
classes can be related by the hasAntenna property.
Antenna A class that represents the rfid antenna(s).
This class represents infrared sensors that for this project it can be used to warn to turn
InfraredSensor on/off some rfid sensor where some presence is detected. For that case the InfraredSen-
sor class controls the RfidSensor.
This is an equivalent class that is “activated” when the data property detectedPresence is
TurnOnInfraredSensor
true.
This one represents object materials. This class have two subclasses Dangerous or Sim-
Material
ple, it’s just to mark the material type that can influence at the moving time.
Dangerous This class represents dangerous materials (e.g. chemical).
Simple This class represents simple materials (e.g. plastic).
As the AllowedPerson and AllowedPlace, this is an equivalent class that represents if the
AllowedMaterial
object is of a simple material (using ofMaterial property).
2.1. Merging I2oTology and Iot-Lite
The Figure 1 shows the merge between I2oTology and IoT-Lite. Since I2oTology is a
tracking-oriented ontology, IoT-Lite’s Object class does not apply to it because it charac-
terizes the family of objects that aren’t part of a Device, such as a desk and a chair. Objects
like these in I2oTology receive an RFID code and are therefore treated as TagDevice. For
the location implementation strategy, the Laboratory and ClassRoom classes were cre-
ated, these are subclasses of Place and because of this, it was not necessary to use the
Deployment class, which in the end, was replicating information that those classes store.
The hasLocation and hasAttribute properties were adapted for the project, so they didn’t
take IoT-Lite in the nomenclature once they were changed. The Teacher, Functionary
and Student classes were added to represent and classify people, these are subclasses of
Person class and participate in the traking process to define who is moving the device.
The AllowedPlace, AllowedPerson AllowedMaterial, and TurnOnRfidSensor classes
are equivalent classes used to “trigger some event”. It is from them that the ontology will
give suggestions and confirmations to the System used to test it.
The Material class is used only to define the material that the device is made of,
it may be Dangerous or Simple types. As subclasses of the SensingDevice, RfidSensor
and InfraredSensor have been added, which are the types of sensors that this System
communicates. An RfidSensor can have multiple antennas, this is why the Antenna
1
Equivalent classes are necessary conditions. When an individual follows exactly this conditions, it’ll
be inferred that this is an individual of this class.
Figure 1. The I2oTology
class of RfidSensorCharacteristic was added.
2.2. How it works
Here is a simple example showing how is the inference of I2oTology (ilustration at Fig-
ure 2). Assuming the following registered individuals: obj1, func1 (is a Functionary),
stud1 (is a Student), plastic (is a Simple material) and room1 (is a Laboratory). The
obj1 has a deviceTag (a data property added) that is “f3h532w”, it canBeIn some Lab-
oratory, canBeMovedBy some Functionary and is ofMaterial plastic. This object is-
MovedBy stud1 and right now hasLocation room1. If the reasoner is started, there will be
the following inferences:
• obj1 is a TagDevice because it has a deviceTag;
• obj1 is individual of AllowedMaterial because it’s of a simple material;
• obj1 is individual of AllowedPlace because it canBeIn some Laboratory and
room1 is a Laboratory.
The obj1 isn’t individual of AllowedPerson because it canBeMovedBy some
Functionary and stud1 is a Student. The ontology and it’s inferences in this context
“answer” to the system (explained in next section) the SO status. The status information
can show, for example, with who is the SO? Where is it right now? Can this person move
this object? Can this object be in this place? With this information, the system can act
when something wrong occurs.
3. Implementing the ontology
As said, the ontology presented was partially implemented in a web system to be tested.
The project aims to implement a tracking system to improve the object verifications at the
University. The system uses RESTful for server communication. To implement the ontol-
ogy, it was created an ontology service layer to build a system-ontology communication.
Figure 2. I2oTology Inference Example
To develop the ontology API to work with the I2oTology, it was used the Jena Framework
[Jena 2007] for building Semantic Web and Linked Data applications.
To make a simple test with the I2oTology, some SO’s were registered in both
system’s database and ontology. The web system has a page to insert some RFID tags
read by a RFID reader. The user can access this page by some palm computer which is
connected to RFID reader, read the tags and insert in the text area and press ”send”. After
the user press ”send”, the system will verify if this object is registered in the database and
then will send all the read tags to the ontology service (also with the actual location where
all tags were read) to make all needed verifications and inferences. After some seconds,
the ontology give the system some results, as well as if the SO is in the right place with
the right person and the system prints each Smart Object result to the user. The system
idea is to monitor each movement and activity of each object, knowing at all the time
where it is, with who it is and if it can be with that person in this place, and to act in any
way every time something wrong happens.
4. Conclusion and Future Work
This paper presented a tracking-oriented ontology, the I2oTology, which aims to be im-
plemented in UESC University and in the question of tracking, it can be said that it can
be applied to similar contexts. This ontology was partially tested with a system which is
still being developed. This test consisted to evaluate the ontology response about which
room is an Smart Object, who is moving it and if it can be with this person at this place.
There are others classes, properties and situations to be tested with the I2oTology and this
is some of the future work topics. Some of next steps is to implement a way to use the
geo:Point class to track the exact place where the SO’s are located (with latitude, longi-
tude and altitude information), to integrate the system registry page with the ontology to
create the SO in the database and in the ontology maintaining the data consistency and to
know how much the ontology is right so with this results it’ll be possible to be sure about
the ontology certain.
References
Agarwal, R., Fernandez, D. G., Elsaleh, T., Gyrard, A., Lanza, J., Sanchez, L., Geor-
gantas, N., and Issarny, V. (2016). Unified iot ontology to enable interoperability and
federation of testbeds. In Internet of Things (WF-IoT), 2016 IEEE 3rd World Forum
on, pages 70–75. IEEE.
Bermudez-Edo, M., Elsaleh, T., Barnaghi, P., and Taylor, K. (2016). Iot-lite: a lightweight
semantic model for the internet of things. In Ubiquitous Intelligence & Comput-
ing, Advanced and Trusted Computing, Scalable Computing and Communications,
Cloud and Big Data Computing, Internet of People, and Smart World Congress
(UIC/ATC/ScalCom/CBDCom/IoP/SmartWorld), 2016 Intl IEEE Conferences, pages
90–97. IEEE.
Brickley, D. (2006). Basic geo (wgs84 lat/long) vocabulary. Documento informal escrito
en colaboración.
Chandrasekaran, B., Josephson, J. R., and Benjamins, V. R. (1999). What are ontologies,
and why do we need them? IEEE Intelligent Systems and their applications, 14(1):20–
26.
Compton, M., Barnaghi, P., Bermudez, L., Garcı́A-Castro, R., Corcho, O., Cox, S., Gray-
beal, J., Hauswirth, M., Henson, C., Herzog, A., et al. (2012). The ssn ontology of the
w3c semantic sensor network incubator group. Web semantics: science, services and
agents on the World Wide Web, 17:25–32.
Gil, J. M., Montes, J. F. A., Alba, E., and Aldana-Montes, J. (2018). Optimizing ontology
alignments by using genetic algorithms.
Jena, A. (2007). semantic web framework for java.
Keat, N. W. and Shahrir, M. S. (2017). Ontology for customer centric digital services and
analytics. In AIP Conference Proceedings, volume 1905, page 040024. AIP Publish-
ing.
Khodadadi, F. and Sinnott, R. O. (2017). A semantic-aware framework for service defi-
nition and discovery in the internet of things using coap. Procedia Computer Science,
113:146–153.
Li, J., Yan, Q., and Chang, V. (2018). Internet of things: Security and privacy in a
connected world.
McGuinness, D. L., Van Harmelen, F., et al. (2004). Owl web ontology language
overview. W3C recommendation, 10(10):2004.
Pease, S. G., Conway, P. P., and West, A. A. (2017). Hybrid tof and rssi real-time semantic
tracking with an adaptive industrial internet of things architecture. Journal of Network
and Computer Applications, 99:98–109.
Sene Júnior, I. G., Bulcão-Neto, R. d. F., et al. (2014). Anatomy of a semantic con-
text interpreter with real-time events notification support. In Proceedings of the 20th
Brazilian Symposium on Multimedia and the Web, pages 159–162. ACM.
Veiga, E. F., Arruda, M. F., Neto, J. A. B., and Bulcão-Neto, R. d. F. (2017). An ontology-
based representation service of context information for the internet of things. In Pro-
ceedings of the 23rd Brazillian Symposium on Multimedia and the Web, pages 301–
308. ACM.