Proposal of a Standard Vocabulary for Services Discovery on the
Internet of Things
Mayka de Souza Lima, Admilson de Ribamar L. Riberio and Edward David Moreno
Computing Department, Segipe Federal University, Aracaju, Brazil
Keywords: Internet of Things, Web of Things, Vocabulary, Semantic, Mark-up, IoT, Ontology.
Abstract: Internet of Things (IoT) is the paradigm that will dominate the computing world in the coming years. Thus,
studies should be performed in order to ensure improvements in the search for technologies that can create
new standardized semantic vocabularies to find services provided by IoT. The aim of this work is to
understand the concepts of vocabularies of ontology as well as technologies that use such semantic
vocabularies. Thus, it was concluded that the proposed vocabulary increase the chances of finding the IoT
services, integrated with applications that work with many wireless sensor devices.
1 INTRODUCTION
With the exponential growth of web business
investment, the internet has become one of the main
existing business channels nowadays. Soon the
development of technologies to create internet
access platforms, faster and faster, efficient and
cheap generated an unprecedented technological and
social transformation.
Connect virtually any device to the Internet
facilitates not only the design of intelligent
buildings, but also makes possible remote
monitoring of other devices from water energy and
meters, environmental sensors, to medical implants.
This interaction with other devices raises a paradigm
that by 2025, according to forecast NIC (US
National Intelligence Council) will dominate the
computing world, called Internet of Things (IoT).
In short, the IoT is the diffuse presence of a
variety of things or objects around us, for example,
RFID tags - Radio Frequency Identification, smart
mobile phones, wireless sensor networks - WSN,
between others, that communicate by exchanging
many messages, even by simple sensors (Atzori,
2010).
Some projects are directed to metadata
architectures production, but are made to take into
account three aspects of interoperability: semantic,
enabling the understanding of the meaning of each
element of the described feature, along with the
associations found in it, making sure the use of
specific vocabularies, ontologies and metadata
standards that are essential; syntactic, determining
how metadata should be coded to the information
transfer, using technology employed as the XML
(eXtensible Markup Language) language; and
structural, that specifies how resources are
organized, along with the types involved and the
possible values for each type (Perera, 2013).
Therefore, a document or a file where they are
formally defined the relationships between concepts
in the semantic Web, is called ontology. A
taxonomy formed classes and subclasses of objects
related to each other more a set of inference rules
that can use language like DAML (DARPA mark-up
language agent) developed as an inference based on
RDF (Resource Description Framework).
Create or edit an ontology for semantic
vocabulary can be accomplished through a tool
called Protégé-2000. Widespread for the semantic
Web, this tool allows to define vocabularies in
different microformats as, for example RDF and
RDF Schema. Highly customizable, allows the
conceptual modeling of these languages, making it
possible to create a standard vocabulary to discover
services used by the IoT.
This work aims to propose a standard
vocabulary, creating a necessary ontology for
services used by the IoT, demonstrating its
functioning and interaction with the human life
reality.
The remainder of this paper is organized as
follows. In section 2, will be shown on existing
requirements nowadays and contribute to the
129
de Souza Lima M., de Ribamar L. Riberio A. and David Moreno E..
Proposal of a Standard Vocabulary for Services Discovery on the Internet of Things.
DOI: 10.5220/0005493001290134
In Proceedings of the 11th International Conference on Web Information Systems and Technologies (WEBIST-2015), pages 129-134
ISBN: 978-989-758-106-9
Copyright
c
2015 SCITEPRESS (Science and Technology Publications, Lda.)
creation of a semantic vocabulary. Section 3
describes the proposed semantic vocabulary that will
be explained and possible tool that can be used in
creating it. Finally, in section 4, the work related to
the subject covered in this paper and its relation to
the ontology and semantic. Conclusion and
references are given in section 5.
2 INTERNET OF THINGS
VOCABULARY
REQUIREMENTS
The semantic vocabularies used by the Internet of
Things, have basic requirements needed to extract
the communication results with the various services
or applications. These basic requirements are
described in this paper, to name a few of the
concepts that could be used to serve as the basis for
creating a standard vocabulary, such as the language
syntax, ontology and semantic markup.
2.1 Syntax
Through this feature the syntax of services provides
interoperability, allowing the automation of the use
of information based on their semantics. Thus, to
describe things, an unique to identify them, is
necessary both in the current Web and Semantic
Web, which are called features and identified using
URIs (Uniform Resource Identifier).
The syntax description, refers to the XML
standards (Extensible Markup Language), XML
namespace and XML schema (W3C, 2014). The
idea is to associate a namespace URI and even if a
name appears in more than one space, must be
unique in its namespace. These standards describe a
class of objects called XML documents and the
behavior of computer applications that process these
documents.
Being an open standard (non-owner) and simple
to read and write by software applications, a
document in XML format makes it an excellent
format to exchange data between different
applications.
2.2 Ontology
Ontologies are key requirements to create a semantic
vocabulary. They serve as support for the
description, publication, discovery and Web service
composition being: OWL-S (Ontology Web
Language Semantic); OWL-Time, resources
ontology and domains ontology.
The ontology language OWL-S is semantic
markup and proposes an ontology language for Web
services for general purpose, able to describe any
type of Web services regardless of the domain it
belongs to the service.
OWL-Time defines a higher level of temporal
concepts and contains a specification of the time
theory required for semantic Web applications. Is
used to describe temporal content Web pages, as
well as describe properties and web services time
constraints.
Ontologies resources proposes a description of
Web services and their intention is to be used in
conjunction with OWL-S. And the domain is used to
describe of Web services in order to provide the
semantics necessary to enable automatic discovery
and composition (Hachem, 2011).
2.3 Semantic Markup
The patterns of semantic annotations that emerged
by the W3C, were: RDF, RDFa, Microformats and
Microdata. The semantic annotation is the process
by which includes semantic information or metadata
and add machine-readable information to existing
content.
Thus, if the marking is ready to find the most
relevant result, the semantic annotation will add
diversity and richness to the process. With
Microformats, which are not a new language but an
idea to solve the simplest problems of semantic
Web, by today's standards and use in semantic
markup, it is easy to use, with the most used formats,
hCalendar (for publication events) and hCard (for
people, companies and organizations in general).
The RDFS, XML-based language, considered a
comprehensive language, allowing you to define
ontologies with her. But, being a complex language
is not feasible to use it to mark Web pages
semantically. The RDFa language (RDF - in -
attributes) is the standard semantic annotation that
does not have the need of using new markup
elements, being performed by corporate attributes on
any element (Norway, 2013).
The microdata are considered the latest markup
language and competitor of current standards of the
Semantic Web annotations. Heavily influenced by
microformats, adopting a range of attributes such as
hCard, hCalendar and other microformats.
WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies
130
3 PROPOSAL FOR AN IOT
VOCABULARY
The initial proposal will be creating a standard
semantic vocabulary to obtain a better result to
discover services used by the IoT.
Then, we use a tool called Protégé, to develop
new semantic vocabulary to be tested with
applications or devices that work in the IoT.
Correlate the Web services more robust and
effective manner, and to facilitate the search of users
in the use of information in the applications with the
IoT is the main purpose of this proposed paper.
With this platform, developed by the Stanford
Medical Informatics research group at Stanford
University medical school, will allow the ontology
construction, namely the vocabulary to be used in
the discovery of services in the IoT.
Based on the Java language, the Protégé is
extensible and deals with a strong community of
developers, academics and users, who use it to
knowledge solutions in various areas such as
biomedicine. Chosen as the semantic creation tool
due to several factors: is a publisher of open source
ontologies, supports editing Web client or via
desktop client, access can be made via the Web or
application installed on the machine, allowing the
development of ontologies in various formats such
as: OWL; RDF and XML Schema.
It also allows the use of existing vocabularies
when creating the ontology itself and the export of
ontology to the desired format, in addition to having
a graphical interface easier to handle, according
(Horridge, 2014).
The graphical interface provides access to the
menu and toolbar bar, and presently five viewing
areas (views) which act as navigation modules and
editing classes, attribute, forms, forums and search
the database of knowledge, providing the data input
and retrieval of information. The Protégé platform
interface can be viewed in Figure 1. According to a
pattern to be used can be constructed by the
ontology vocabulary defined by this tool. Thus,
describe the classes, subclasses, instance values and
capable of interacting with the applications used in
Web to IoT.
The vocabulary to be created should follow the
standards previously defined in order to have more
interaction with IoT applications and embedded
devices. The whole structure to be mounted must
have open-source systems, devices that are
connected to the Internet of Things, and a Web
interface to bring information services found by IoT.
Decrease the amount of human intervention plans in
search of information as faster the semantic analysis
and run the commands requested by users with
better understanding are the metrics needed to get an
effective result in the discovery of services
connected to the IoT applications in various devices.
Figure 1: Protegé platform interface.
ProposalofaStandardVocabularyforServicesDiscoveryontheInternetofThings
131
The Web Services Description Language
(WSDL) can describe a service by specifying the
location and describing the operations provided for
him. All documents are in XML and provide enough
information to interact with the Web service. Have
as elements to define the service interface: types,
XML schemas; messages describing details of the
methods and their parameters; type of port that
defines operations. Allowing its interaction with
other systems that identify the service.
A tool to be used for the discovery of services in
WSDL will be the Lumina plugin. This ontology
concepts associated with the input parameters, so
that the search system according to see description
to be made by the reference model (Menegazzo,
2009). Lumina is an eclipse plugin that allows to
discover Web services based on semantic built. It
has a graphical user interface and its repository
services is UDDI (Universal Description, Discovery
and Integration).
An example of the use of semantic vocabulary
would be to use some client-side objects with a
service by doing a proxy class that "mimic" the web
service method calls. Working with the proxy
instead of writing SOAP messages (Simple Object
Access Protocol) directly. The proxy class manages
the building, sending and receiving SOAP messages
and OWL-S language is intended to enable the
automation of discovery, invocation and Web
Service composition through semantic descriptions.
The axioms of OWL-S are defined in OWL and
OWL-S has descriptions of classes, sub-class
hierarchy and definitions of types and relationships
between classes. The description of the functionality
of a service is referenced by a profile of OWL-S
ontology externally with the concepts defined in
OWL ontology in the domain that owns the Web
Service in question.
Analyzing the available work in this area, we can
verify that applications involving IoT and semantic
vocabularies in search of services need to get to an
improvement in the syntax and semantics of
communication between them, creating new
vocabularies that can interact with different devices
connected to things.
4 RELATED WORK
In this section will be elucidated major systems that
use semantic vocabulary for their services and their
characteristics. These features are related to the
requirements that have been seen before.
4.1 Steer
Computer program implemented to support
mechanisms for service discovery, as well as tools
and semantic Web services.
STEER compound (Task execution Semantic
Editor) and PIPE (Pervasive Instance Provision
Environment). A Web client with a Java interface
classifies services based on their input data and
semantic outputs. Through a GUI manager
(Graphical User Interface) PIPE services are
facilitated for the users so that in this way can
perform Web services for common tasks. Using as
ontology language, OWL, as a powerful modeling,
naturally enough to describe many areas. (Masuoka,
2003).
4.2 Middleware IoT
Middleware architecture for it consists of three
modules discovery, composition and estimation, and
knowledge base. This application performs a request
or a service by making detection, or a composition
for estimation of the manipulated object. To solve a
request, this module accesses the module (discovery)
and makes the knowledge base by performing the
creation of a composition of services (
Zhexuan,
2010). The result is obtained by running this
composition within the existing network.
4.3 Ontonym
Invasive system which has the need to interpret large
amounts of data from many sources. Developed to
address the temporal properties of the data context
as a requirement. Having as a criterion to function
using a set of ontologies to represent the
fundamental concepts in pervasive computing
(Stevenson, 2009).
The concepts that surround it to carry out a
comparison with the ontology of pervasive
computing are: time, using the ontology language
OWL-Time; location, generated by a GPS; people,
where personal and social networking information is
collected; and the ontological sensors generating
data.
The works related are limited in specific services
for the use an ontology as OWL and OWL-S. The
profile and the process model of these services
specify Web service features that can be used by
agents for semantic discovery, invocation and
composition services.
With an ontology class, you can define how the
service will be mapped from abstract definitions of
WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies
132
the profile and the process for specific information
to be made by means of messages exchanged
between the user and application. Some of these
applications, using traditional architectures for Web
services, such as systems for wireless devices with
the set of protocols UPnP (Universal Plug and Play)
(Masuoka, 2003). However, OWL-S does not
require the use of service oriented architectures and
several studies reported in the literature using OWL-
S based on other architectures.
Map the abstract elements of OWL-S of the
evidence in WSDL (Web Services Description
Language), and reuse all the existing infrastructure
for Web services, maintaining that the Semantic
Web should be an extension of the Current Web,
leads to an advantage to locate Web services in the
Semantic Web layers.
4.4 CoAP Framework
The COAP protocol (Constrained Application
Protocol) devices for handling resources with a
REST framework and is used in generic browsers
for IoT. This protocol adopts HTTP resources
(Hypertext Transfer Protocol) allowing an evolution
of the Web from a simple document retrieval
mechanism by GET REST methods, POST, and
DELETE, integrating the URIs in the browser
(Kovatsch, 2013).
From this interaction arise APIs (interactive
programs) easy navigation of users on the Web
through links, thus discovering new devices or
services.
5 CONCLUSIONS
The vocabulary to be created will primary to provide
as a benefit, the most effective discovery of the
various services that are used by the Internet of
Things. So, returning to the applications more
concise information, making it easier to use between
users and the various things interconnected to the
Internet.
Once created, with the defined ontology
language, this vocabulary will help the semantic
Web of several applications that work-related RSSP
networks and the Internet of Things. Therefore, the
focus is the creation of standard vocabulary for the
IoT, leaving all the attention to be directed to the
main goal of the research.
For future work, there is the possibility to
develop applications that use this semantic
vocabulary demonstrating their interaction with
other devices available on the Internet of Things
using different types of ontologies, syntax and
semantics.
REFERENCES
Atzori, L., Iera, A., Morabito, G, 2010. The Internet Of
Things: A Survey. Italia.
http://Www.Science.Smith.Edu/~Jcardell/Courses/Egr328/
Readings /Iot%20survey.Pdf last accessed September
15, 2013.
Hachem, S., Teixeira, T., Issarny, V., 2011. Ontologies
For The Internet Of Things. Inria Paris-Rocquencourt.
Hal Id: Hal-00642193. https://Hal.Inria.Fr/Hal-
00642193/Document last accessed January 01, 2015.
Horridge, M., 2014. Protégé. Manchester.
http://owl.cs.manchester.ac.uk/publications/talks-and-
tutorials/protg-owl-tutorial/ last accessed June 24,
2014.
Kovatsch, M., 2013. CoAP for the Web of Things: From
Tiny Resource-constrained Devices to the Web th
Internacional Workshop on the Web of Things.
UbiComp’13 Adjunct. Zurich, Switzerland.
http://www.vs.inf.ethz.ch/publ/papers/mkovatsc-2013-
wot-copper.pdf last accessed March 15, 2015.
Masuoka, R., et al, 2013. Ontology – Enabled Pervasive
Computing Applications. IEEE. University Of
Maryland.
http://masuoka.net/Ryusuke/papers/20030915-Task-
Computing-IEEE-Intelligent-Systems-September-
October-2003.pdf last accessed January 05, 2015.
Menegazzo, C., Oliveira, D., Claro, D. B, 2009. Uma
análise conceitual das linguagens semânticas de
serviços Web: Comparação entre OWL-S, WSMO e
SAWSDL. UFBA – Universidade Federal da Bahia.
http://homes.dcc.ufba.br/~dclaro/download/IADIS_CI
AWI_2009_CAMERAREADY.pdf last accessed
March 10, 2015.
Norway, F., 2013. Semantic Markup Report
Microformats, Rdfa, Grddl, Microdata and Odp. Nce
Tourism.
http://www.vestforsk.no/filearchive/semantic_markup
_report.pdf last accessed October 15, 2014.
Perera, C., Zaslaysky, A., Christen, P., 2013. Context
Aware Computing For The Internet Of Things: A
Survey. Ieee Communications Surveys & Tutorials
Journal. Http://Arxiv.Org/Abs/1305.0982 last
accessed May 18, 2014.
Stevenson, G., et al, 2009. Ontonym: A Collection Of
Upper Ontologies or Developing Pervasive Systems.
CIAO, Proceedings of the 1st Workshop on Context,
Information and Ontologies.
http://dl.acm.org/citation.cfm?id=1552271 last accessed
December 15, 2014.
W3C, 2014. OWL Web Ontology Language Overview.
Http://Www.W3.Org/Standards/Semanticweb/Ontolog
y last accessed December 15, 2014.
ProposalofaStandardVocabularyforServicesDiscoveryontheInternetofThings
133
Zhexuan, S., Cardenas, A., Masuoka, R., 2010. Semantic
Middleware For The Internet Of Things. IEEE. Fujitsu
Laboratories Of America, Inc. Http://Www.Flacp.
Fujitsulabs.Com/~Cardenas/Papers/Iot2010.Pdf last
accessed June 19, 2014.
WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies
134
