Semantic Sky: A Platform for Cloud Service Integration based on Semantic Web Technologies

Abstract

These days, the number of data sources an ordinary computer user works with every day is very large and continues to grow. With the increasing number of cloud services with specialized functionalities, the users are faced with the necessity to routinely perform manual actions to interchange data among different cloud and web services, in order to perform more complex and composite actions. These actions always require a certain amount of dedicated time from the user, who has to change the context in which he work, in order to take the actions and transfer data from one system to another. In this paper, we present a software platform, based on the concepts and technologies of the Semantic Web, which provides the users with a unified and simple composite approach to the different services they use, and crates a simple flow of information from one infrastructure to another. The system is able to automatically discover the context in which the user is working, and offer him the actions which can be used over the data within the context. In this way, the user can completely focus on his tasks in his work environment, and get relevant information and possible actions in that context. This system automates the execution of the users’ tasks, which leads to improvements in their productivity, information exchange and efficiency. The system is called “Semantic Sky” and represents a platform where many cloud services are interconnected by the use of semantic web technologies.

Full text

Semantic Sky: A Platform for Cloud Service Integration
based on Semantic Web Technologies
Dimitar Trajanov
Riste Stojanov
Milos Jovanovik
Faculty of Computer Science and
Engineering
Faculty of Computer Science and
Engineering
Faculty of Computer Science and
Engineering
Skopje, Macedonia
dimitar.trajanov@finki.ukim.mk
Skopje, Macedonia
riste.stojanov@finki.ukim.mk
Skopje, Macedonia
milos.jovanovik@finki.ukim.mk
Vladimir Zdraveski
Petar Ristoski
Marjan Georgiev
Faculty of Computer Science
and Engineering
Faculty of Computer Science and
Engineering
Faculty of Computer Science and
Engineering
Skopje, Macedonia
vladimir.zdraveski@finki.ukim.mk
Skopje, Macedonia
petar.ristoski88@gmail.com
Skopje, Macedonia
marjan.georgiev@gmail.com
Sonja Filiposka
Faculty of Computer Science
and Engineering
Skopje, Macedonia
sonja.filiposka@finki.ukim.mk
ABSTRACT
These days, the number of data sources an ordinary computer
user works with every day is very large and continues to grow.
With the increasing number of cloud services with specialized
functionalities, the users are faced with the necessity to
routinely perform manual actions to interchange data among
different cl oud and web services, in order to perform more
complex and composite actions. These actions always require a
certain amount of dedicated time from the user, who has to
change the context in which he work, in order to take the
actions and transfer data from one system to another. In this
paper, we present a software platform, based on the concepts
and technologies of the Semantic Web, which provides the
users with a unified and simple composite approach to the
different services they use, and crates a simple flow of
information from one infrastructure to another. The system is
able to automatically discover the context in which the user is
working, and offer him the actions which can be used over the
data within the context. In this way, the user can completely
focus on his tasks in his work environment, and get relevant
information and possible actions in that context. This system
DXWRPDWHV WKH H[HFXWLRQ RI WKH XVHUV¶ WDVNV ZKLFK OHDGV WR
improvements in their productivity, information exchange and
efficiency. The system is called ³6HPDQWLF Sky´DQGrepresents
a platform where many cloud services are interconnected by the
use of semantic web technologies.
Categories and Subject Descriptors
H.3.3 [Information Storage and Retrieval]: Information
Search and Retrieval - information filtering, retrieval models,
search process
General Terms
Algorithms, Performance, Design
Keywords
Semantic Web Technologies, Cloud Services, Named
Entity Extraction, Automation of Processes.
1. INTRODUCTION
The information we work with every day is obtained from
different sources ± email, Facebook, Twitter, local documents,
enterprise services, etc. Depending on that information, we
usually take some actions (e.g. we share them, add them into a
³Wo-do´ list, etc.). In the last years, with the increasing number
of cloud services [5] with specialized functionalities, we came
across the need to routinely perform manual actions to
interchange data among these cloud services, in order to
perform more complex and composite actions. These actions
always require a certain amount of dedicate d time from the
user, who has to change the context in which he works
manually, in order to take the actions and transfer data from one
system to another. Furthermore, we often need mechanisms for
detection and description of the entities which are found within
texts and information that we get from different services.
With the use of the technologies of the Semantic Web [18] in
this approach, we can automate some of these tasks and
processes, and eliminate the manual work. Even though the
main purpose of the Semantic Web is to give meaning to the
resources on the World Wide Web, semantic annotation of other
resources can be used to automate some of the processes
involved in the everyday service consuming flow.
Permission to make digital or hard copies of all or part of this work for
persona l or cla ssroo m use is granted without fee pr ovided that copies are
not mad e or distributed for profit or commercial advanta ge and that
copies bear this notice and the full citation on the first page. To copy
otherwise, or republ ish, to post on servers or to redi stribu te to lists,
requires prior specific permission and/or a fee.
I-Semantics 2012, 8th Int. Conf. on Semantic Systems, Sept. 5-7, 2012,
Graz, Austr ia.
Copyright 2012 ACM 978-1-4503-1112-«
109

In this paper we present a software platform, developed using
the technologies of the Semantic Web, which provides the users
with a unified and simple composite approach to the different
services they use, and with a simple flow of information from
one infrastructure to another. The system is able to
automatically discover the context in which the user is working,
and offer them the actions that can be used within the context.
The users can completely focus on their tasks in their work
environment, and get relevant information and possible actions
in that context. This system automates the execution of the
XVHUV¶WDVNVZKLFKOHDGVWRLPSURYHPHQWVLQWKHir productivity,
information exchange and efficiency.
2. RELATED WORK
The potential and influence that the Cloud systems have had
over the development and use of information technologies
during the last years, attracts more researchers to work in this
area. One of the challenging aspects in this area is the
integration of semantic technologies that should enable drastic
changes in the opportunities for application integration with the
user environment and services [15][21][26][29]. Today, this
research fiel d is very active, with many projects aiming towards
solving these existing problems. As an example in terms of
development of desktop applications which use semantic
technologies, the project Semantic Desktop [28] gives a
proposal for layered architecture based on ontologies for
different levels. By organizing the data and annotating the
resources, which means creating associations between resources
and files based on extraction of ke ywords and relations, and
analyzing the history of use, it is possible for complex actions to
be related to the context in which the user is working at the
moment, and to be offered to him. The described project aims
towards the integration of semantic technologies in desktop
environments, but excludes the integration of many different
Cloud infrastructures.
On the other hand, the project mOSAIC (Open-Source API and
Platform for Multiple Clouds) [20], intents to build an
application platform that will integrate many Cloud services in
order to answer complex user requirements. This project is
under way and will provide a framework for development of
applications that does not depend on the infrastructure they will
be executed on. This framework will use a Cloud ontology and
semantic representation of the resources [23]. An application
will be able to define its own requirements and the decision
about the place of execution will depend on them. The
H[HFXWLRQZLOOEHGULYHQE\FHUWDLQ³DJHQWDSSOLFDWLRQV´
The project SITIO [12] is another research project that
integrates many Cloud services, but unlike mOSAIC, the
integration here is done only at the service level. The system
implemented here enables different developers to create
applications and integrate them in their own system. Several
ontologies for semantic annotation of the available services, as
well as certain tools for semi-aut omated annotation of their
services, are available to the developers. Both mentioned
projects are mainly aiming at the issues of interconnecting
different Cloud infrastructures and defining interfaces for their
mutual communication.
Anzo Data Collaboration Server lies at the heart of the
Cambridge Semantics architecture [4]. The Open Anzo project
includes an open source enterprise-featured RDF quad store,
coupled with a unique semantic middleware platform. Together
they provide the developers with a host of features necessary for
the creation of sophisticated semantic technology standard
(RDF, OWL, SPARQL) grounded applications. Features
include support for multiple users, distributed cl ients and
services, offline work, real-time notifications, named-graph
modularization, versioning, access controls, and transactions
with preconditions.
Babylon±Enterprise [6] is an enterprise information retrieval
system, which represents a web-configured client-server
system, based on a Windows client installed on the end -XVHU¶V
workstation and an enterprise application server. All of the
employees in the enterprise can easily access all company
information from their work environment.
OntoBroker is an ontology repository from ontoprise, which
includes a high performance deductive reasoning engine [2][3].
Reasoning with rules is a major unique selling point for
ontoprise. OntoBroker integrates a connector framework which
makes it easy to connect a multitude of data sources like
databases, web services, etc. Although it combines structured
and unstructured data in one framework, OntoBroker is easy to
extend and to integrate into existing IT landscapes and
applications, as it offers a variety of open interfaces.
OntoBroker i s also closely connecWHG WR RQWRSULVH¶V RQWRORJ\
modeling environment OntoStudio, which is the development
environment for handling ontologies, mappings to information
sources, rules, generating queries, creating business intelligence
reports, etc. [2]. For many customers OntoBroker serves as a
common semantic layer which is accessed by various
applications and integrates different information sources.
None of the systems described above fully automates the
process of user context extraction and invocation of relevant
actions. These systems have some semantic features, but none
of them integrates a context recognition capability, automatic
action proposition, automatic action execution and auto-
generated API interface, which makes them require a
customization for every future extension, in terms of both the
knowledgebase or and a client-side application.
3. SOLUTION DESCRIPTION
The system we developed LVFDOOHG³6HPDQWLF 6N\´EHFDXVHLW
is an environment where many cloud services are
interconnected by the use of semantic web technologies. A
detailed description of the solution follows.
3.1. Architecture
The core of the system is developed with the Play MVC
framework [27]. The system architecture, shown on Figure 1,
consists of several components, listed below.
The user interaction with Semantic Sky happens through
Application plug-ins. Any third party application, which
implements the Semantic Sky API, can be integrated as a plug-
in into the system. These plug-ins extract the user's working
context and, when possible, infer additional metadata. This
information is then sent to the Context Extraction Module. In
the current implementation of the system, plug-ins for Gmail,
Microsoft Office Word and Microsoft Office Outlook are used
(see Section 3.3).
The Knowledgebase contains all of the data used by the user¶V
applications. Our system extracts all of the information from it.
New data can be easily added to the Knowl edgebase, either by
uploading a new OWL/RDF file, by adding a link to an
OWL/RDF file or by adding an URL to a SPARQL endpoint.
After the data is added to the Knowledgebase, it is indexed and
ready to use.
The Context Extractor Module, shown on Figure 2, is the entry
point to the core of the solution. It accepts the text-based data
sent from the Application plug-ins. For each token (word) in the
110

text, the system extracts all resources related to it, using an
index previously created in the Knowledgebase. All entities
retrieved from the search are converted into semantic resources,
and grouped by their type. This module uses a modification of
the Apache Lucene Core engine [14], adapted to fit our
solution. The modifications included using B+ hash trees to
speed up the on-disk RDF lookups, as proposed in [24].
The Semantic Web Services Repository contains semantically
annotated SOAP and RESTful web services, which will be used
later to build actions for the user to execute. The system
provides a simple form for annotation [8] and uploading of new
web services. The SOAP web services are annotated using the
Semantic Annotations for WSDL and XML Schema
(SAWSDL) framework [10]. Although there are several
schemas for semantic annotation of RESTful services [19],
most notably hRESTS [16] and MicroWSMO [17], we use a
custom schema. The existing schemas are more robust than
what we need, so we decided to go with a custom, more light-
weight approach, which fits better with our solution. Our
schema uses the base-URL of the service, the method type and
the names of the input parameters and the output parameter as
properties which are semantically annotated. Each web service,
regardless of the type, can be annotated with any ontology
within the system.
The Action Search Module, shown on Figure 3, accepts the
previously extracted resources from the Context Extractor
Module, and aligns them as inputs. The system uses a custom
algorithm to search through the SWS Repository for web
services or compositions of web services which can be invoked
with the previously identified inputs. The algorithm detects the
semantic type of the desired output, and scans through the SWS
Repository for services which provide such outputs. The
detected services are tested for the semantic types of their
inputs, in order to select only the services which can be invoked
with the provided inputs. For every input parameter which is
not provided, the algorithm does the same search for services
which can provide it as an output. This is done repeatedly, until
the algorithm comes to a state in which all of the required inputs
for the selected services are provided either from the Context
Extractor Module or as outputs from other services. In this state,
the algorithm has identified one or more candidate web services
or compositions of web services. We call them actions. We rank
the actions using a suitability calculation, based on the number
of services in the composition and the number of para meters
which need to be exchanged at each level of the composition.
As a result, we have actions ordered by relevance to the
extracted resources. To improve the performances of the
module, the actions are stored in cache, structured as an XML
file, which speeds up the search process.
The UI Generator Module, shown in Figure 4, is responsible for
generating the Graphical User Interface (GUI) of the system. In
this module we implemented an algorithm for generic
visualization of different types of resources and entities. The
Figure 1: System Architecture.
Figure 2: Context Extractor Module.
111

algorithm binds an XSLT transformer to each type of resources,
which is used to render that type of resource. Using this
approach, we enable previews of extracted resources, possible
actions, results from the actions, and a preview of the
configuration of the system.
The Module for Action Execution is responsible for the
execution of the semantically annotated actions. This module
receives the RDF resources and the semantic description of the
action as an input. The arguments for the action are provided by
the user. The user selects each input from a group of possible
resources. These resources are the ones returned from t he
Context Extractor Module, grouped by the service input type. If
there is only one resource from the given type, it is
automatically selected. One action can consist of a single
RESTful web service, a single function from a SOAP web
service, or a composition of functions from one or more SOAP
web services. In the former two cases, the invocation is a single
step process. However, for the latter case, we implemented an
algorithm which invokes the functions in a given order and
handles the parameter passing between them. After the
execution of the action, the generated output is displayed to the
user. If the action does not have a visible output, the system
displays a status message. The implementation of this module
uses the Apache Axis2 engine [1].
3.2. Advantages
The main advantage of the system is its capability for
integration with already existing, proprietary and widely used
systems, such as Gmail, Microsoft Office, etc. This provides a
low learning curve for the end users: our system integrates
within their work environment, so they will not have to adapt
into a new, unfamiliar environment. This capability is possible
due to the modular architecture of the solution. The integration
is achieved in few simple configuration steps. First, the user
data should be connected to the system. The data formats can be
relational databases, semantic databases or OWL/RDF files.
This connection process is different from a standard data import
process. Instead, our system indexes the data and stores the
index within the knowledgebase. The next step is a semantic
annotation of the SOAP and RESTful web services, and their
registration in the SWS repository. After this, the system is
ready for use in the new environment.
Another advantage of the system is its scalability. The
architecture allows easy expansion of the number of users, since
Figure 3: Action Search Module.
Fi
g
ure 4: UI Generator Module.
112

the communication with the user environment is service-based.
The modularity of the architecture allows adding and removing
functionalities from the system, in order to conform to the
requirements of the user.
The service-based communication with the user environment
also goes into favor of the overall extensibility of the system. It
can be extended by adding new services, as well as new
ontologies and knowledge. For every new type of data entered
into the system, the use of XSLT makes it is easy to configure a
new UI representation.
Additionally, the system can be used from various types of
application plug-ins, as shown on Figure 5:
x Cloud e-mail plug-ins ± the system can be integrated with
the e-mail client used within an enterprise, for example.
After the integration, the users will get useful information
extracted from each e-mail message they receive, along
with a list of possible action which can be executed over the
identified resources. As we will show later, we developed a
Gmail contextual gadget as a plug-in for the system.
x Desktop clients ± the system can be integrated with various
deskt op applications or it can work on top of all desktop
applications. We developed both Microsoft Office Word
and Outlook add-ins as plug-ins for the system, which are
described in the next section.
x Browser plug-ins ± the system can be integrated with the
web browser used within an enterprise, for example. The
users will get information relevant to their work and a list of
possible actions which can be taken over the data extracted
from the web pages they visit.
x Miscellaneous systems ± Semantic Sky is designed as a
universal system and because of its modular architecture it
can be ext ended and integrated with most of the existing
systems.
3.3. Implementations
In order to test its functionalities and create a working example,
we implemented the Semantic Sky system in our environment
at the Faculty of Computer Science and Engineering in Skopje.
First, we loaded data into the Knowledgebase. We used the
Friend of a Friend (FOAF) ontology [11] to define the basic
relationships among people at the Faculty. In order to use the
relational database from our student e-services system, we
designed a custom ontology which defines all of the concepts
for a faculty environment, matching all the relations from the
existing relational database [22]. The ontology was designed
using the 3URWpJp RQWRORJ\ HGLWRU [25]. We mapped the
relational database with the ontology and deployed a SPARQL
endpoint, using D2RQ server [7] as described in [22]. The
SPARQL endpoint was then connected to the system and the
data it points to was added into the Knowledgebase.
The second step was to register the existing web services into
the SWS Repository. We semantically annotated the services
with the previously created ontologies, using the tool for
annotation provided by the system [8]. Then all of the services
were registered into the SWS Repository.
As an application plug-in, we developed a Gmail contextual
gadget [13]. The plug-in was tested by the faculty staff, in order
to automate the most common tasks within their e-mail inbox.
The plug-in extracts the entities and relations from each
received e-mail message and suggests a list of possible actions
which can be taken over the extracted context. Thus, the users
Figure 5: System integration with user application plug-ins.
Fi
g
ure 6: Gmail contextual
g
ad
g
et UI.
113

are able to quickly understand who the sender of the message is,
what their mutual relation is, and what the task is. Then the user
can automatically execute one or several actions in order to
accomplish the demands from the e-mail.
The Gmail gadget interface is shown on Figure 6. The upper
part lists the entities extracted from the currently opened e-mail
message, grouped by their type. The lower part lists the possible
actions which can be executed over the extracted entities. These
actions can be executed here, directly from the user
environment, saving the users a valuable time. Grading a
student homework assignment, for example, normally requires a
professor to log-in into a different cloud based e-services
system of the Faculty, browse for the course, browse for the
student and then enter the grade. Here, with the use of Semantic
Sky, these steps are reduces to just one ± a single click on a
button within the e-mail message.
Another application plug-in we developed was a Microsoft
Office Word add-in, shown on Figure 8, which analyzes the
contents of the opened document and extract the entities and
relations relevant for the user and his context. The add-in
automatically lists the possible actions for the extracted
resources. These actions can be executed directly from the add -
in, saving valuable time for the user. The main purpose of this
add-in is to assist the professors in the process of reading and
grading student homework assignments. The add-in helps the
users to quickly understand the context of the full document, the
details of the author and the purpose of the document, in order
to help them evaluate and grade the assignment faster and
better.
When the user selects an action from the Gmail or Work plug-
in, an action form is generated. Figure 8 shows an example of a
single action form. The form displays the action name and is
populated with fields for all of the input types necessary for
execution of the action. The user can manually choose the value
for each of the inputs from the list of possible values. In some
action forms, there may additionally be a specific input type,
called User Defined Input, which must be entered manually by
the user. These types of inputs are used when the value of the
specific input type cannot be extracted from the knowledgebase
or the context of the text, but is user dependent. As shown on
Figure 8, the professor needs to enter the grade manually, in
order to grade the student. After he clicks the execute button,
the preexisting web service for grading a student for a given
course ± now semantically annotated within our system ± is
invoked and the grade is entered into the e-services system of
the Faculty.
We also developed a Microsoft Office Outlook add-in, shown
on Figure 9. Much like the Gmail gadget, this add-in analyzes
the content of e-mail messages. The identified entities are listed,
and all of the actions associated with them are displayed in a
drop-down list under the Semantic Sky ribbon. When the user
selects an action, a form similar to the one on Figure 8 is
gener ated on the bottom of the Outlook mail Inspector window,
providing a fast and easy way to execute the appropriate action
within the context of the e-mail message.
4. CONCLUSION
The main goal of this system is the development of the novel
and innovative framework which enables connectivity and
integration, not only of different cloud services, but also of local
data placed on the user machine. The developed system
provides the opportunity for automation of the use of different
services. It also offers an engine which detects and ranks the
Figure 7: Microsoft Office Word add-in.
Fi
g
ure 8: Action Form.
114

possi ble actions which can be executed within a given context,
over the identified concepts.
The system is designed and implemented as a modular
architecture, which allows it to be very flexible and easily
extendable with new modules. It is platform- and technology-
independent, which means the users have no restriction while
implementing application plug-ins for it.
In the process of the development of the system, an efficient
algorithm for automated context extraction from text, which
extracts the most relevant semantic entities and concepts from
the user data, has been developed. This process reduces the time
necessary for data entry when taking an action, because the
entities are already detected and aligned as inputs for the
appropriate action.
Also, an algorithm has been developed for automatic action
suggestion, based on the user context. It includes creating a
composition of web services and automatic execution of the
selected actions. This process can sometimes provide the user
with actions which he is not aware existed within his work
environment, and can help him be more productive. The
automatic detection and composition of services into a single
action means that longer sequences of processes can be
executed with just one interaction from the user, saving him
valuable time.
We also devel oped a methodol ogy for generic visualization of
the extracted entities and concepts, and user interface
generation.
The integration of this solution within everyday systems and
applications, especially enterprise systems, will improve their
usability and integration with other systems and services, and
will improve the overall performance of the users. The features
of the system reduce the time needed by the end users to
discover and take the appropriate action over the data they are
working with within a given context.
This solution aims towards unification of both the knowledge
and the actions which reside within many diverse and isolated
systems, present within the work environment of a common
user.
5. REFERENCES
[1] D. Almaer. Creating Web Services with Apache Axis.
Disponible en, 2002.
[2] J. Angele, M. Erdmann, and D. Wenke. Ontology-Based
Knowledge Management in Aut omotive Engi neering
Scenarios. Ontology Management, pages 245-264, 2008.
[3] J. Angele. OntoBrokerMatur e and approved semantic
middleware. Semantic Web, 2008.
[4] The Anzo Data Collaboration Server: Anzo Architecture,
2012. Revised June 2012, from Cambridge Semantics:
http://www.cambridgesemantics.com/technology/architect
ure
[5] M. Armbrust, A. Fox, R. Griffith, A. Joseph, R. Katz, A.
Konwinski, G. Lee, D. Patterson, A. Rabkin, I. Stoica, A
view of cloud computi ng. Communi cations of the ACM,
53(4):50-58, 2010.
[6] Babylon-Enterprise: Overview, 2011. Revised 2011, from
Babylon Ltd.: http://www.babylon-enterprise.com/
[7] C. Bizer and R. Cyganiak. D2R Server - Publishing
Relational Databases on the Semantic Web. In 5th
International Semantic Web Conference, page 26, 2006.
[8] K. Budinoski, M. Jovanovik, R. Stojanov, and D.
Trajanov. An Application for Semantic Annotation of Web
Services. In 7th International Conference for Informatics
and Information Technology - CIIT 2010. Institute of
Informatics, Skopje, Macedonia, 2010.
[9] T. Erl. Service-Oriented architecture: Concepts,
Technology, and Design. Prentice Hall PTR, 2005.
[10] J. Farrell and H. Lausen. Semantic Annotations for WSDL
and XML Schema. W3c recommendation, 28, 2007.
[11] Friend of a Friend (FOAF), April 2011, http://www.foaf-
project.org/
[12] F. Garcia-Sanchez, E. Fernandez-Breis, R. Valencia-
Garcia, E. Jimenez, J. Gomez, J. Torres-Nicno, and D.
Martinez-Maqueda. Adding Semantics to Software-as-a-
Service and Cloud Computing. WSEAS Transactions on
Computers, 9(2):154-163, 2010.
[13] Google Developers: Gmail contextual gadget, 2011,
Revised May 2011, from Google:
https://developers.google.com/google-
apps/gmail/contextual_gadgets
[14] A. Jakarta. Apache Lucene ± A high-performance, full-
featured text search engine library, 2004.
[15] K. Keahey, M. Tsugawa, A. Matsunaga, and J. Fortes. Sky
Computing. Internet Computing, IEEE, 13(5):43-51, 2009.
[16] J. Kopecky, K. Gomadam, and T. Vitvar. hRESTS: An
HTML Microformat for Describing RESTful Web
Services. In Web Intelligence and Intelligent Agent
Technology, 2008. WI-IAT'08. IEEE/WIC/ACM
International Conference, IEEE, volume 1, pages 619-625,
2008.
Figure 9: Microsoft Office Outlook add-in.
115

[17] J. Kopecky, T. Vitvar, D. Fensel, and K. Gomadam.
hRESTS & microWSMO. Technical report, available at
http://cms-wg. sti2.org/TR/d12, 2009.
[18] T. B. Lee, J. Hendler, O. Lassila. The Semantic Web.
Scientic American, 284(5):34-43, 2001.
[19] M. Maleshkova, C. Pedrinaci, and J. Domingue.
Supporting the Creation of Semantic RESTful Service
Descriptions. 8th International Semantic Web Conference
(ISWC 2009), 2009.
[20] B. Di Martino, D. Petcu, R. Cossu, P. Goncalves, T.
M_ahr, and M. Loichate. Building a mOSAIC of Clouds.
In Euro-Par 2010 Parallel Processing Workshops, pages
571-578. Springer, 2011.
[21] P. Mika and G. Tummarello. Web Semantics in the
Clouds. Intelligent Systems, IEEE, 23(5):82-87, 2008.
[22] M. Mitrevski, M. Jovanovik, R. Stojanov, and D.
Trajanov. Open University Data. In 9th International
Conference for Informatics and Information Technology.
Faculty of Computer Science and Engineering, Skopje,
Macedonia, 2012.
[23] F. Moscato, R. Aversa, B. Di Martino, D. Petcu, M. Rak,
and S. Venticinque. An Ontology for the Cloud in
mOSAIC. In Cloud Computing: Methodology, System and
Applications, Taylor&Francis Group, to be published,
2009.
[24] M. Ngyen, C. Basca, A. Bernstein, Speeding up on-disk
RDF index lookup using B+Hash Trees. In Proceedings of
The 6th International Workshop On Scalable Semantic Web
Knowledge Base Systems, 2010.
[25] N. Noy, M. Sintek, S. Decker, M. Crubezy, R. Fergerson,
and M. Musen. Creating Semantic Web Contents with
Protege-2000. Intelligent Systems, IEEE, 16(2):60-71,
2001.
[26] D. Petcu, C. Craciun, M. Neagul, S. Pani ca, B. Di Martino,
S. Venticinque, M. Rak, and R. Aversa. Architecturing a
Sky Computing Platform. In Towards a Service-Based
Internet. ServiceWave 2010 Workshops, pages 1-13.
Springer, 2011.
[27] A. Reelsen, Play Framework Cookbook. Packt Publishing
Ltd., 2011.
[28] L. Sauermann, G. Grimnes, M. Kiesel, C. Fluit, H. Maus,
D. Heim, D. Nadeem, B. Horak, and A. Dengel. Semantic
Desktop 2.0: The Gnowsis Experience. The Semantic Web-
ISWC 2006, pages 887-900, 2006.
[29] L. Youse, M. Butrico, and D. Da Silva. Toward a Unified
Ontology of Cloud Computing. In Grid Computing
Environments Workshop, 2008. GCE'08, pages 1-10.
IEEE, 2008.
116