Getting it Fast

An Information Systems Oriented Semantic Web Curriculum

Daniela Giordano and Francesco Maiorana

Department of Electrical, Electronic and Computer Engineering, University of Catania, Viale A. Doria, 6, Catania, Italy

Keywords: Information Systems Applications, Mash-ups, RDF, RDS, OWL, Linked Open Data, Knowledge

Representation and Reasoning, Ontology Engineering, Triplify, SPARQL, Gruff, Jena.

Abstract: The fast growing importance of the semantic web and semantic web applications is demonstrated by the

exponentially growing amount of semantic data produced on the web and by the rise of the linked open data

movement. Besides this strong interest there is a request both from academia and industry for well-prepared

students in order to minimize the training effort needed to prepare them for productive work. This paper

describes the teaching experience of a Master's course entitled “Laboratory in software design and

development – semantic technologies”. A detailed description of the curriculum, the rationale underlying

the choice of content and the software tools used, as well as the main lessons learnt from the experience, are

presented.

1 INTRODUCTION

The enormous impact of the World Wide Web

(WWW) is hindered by some problems related to:

 data representation: all the data on the WWW

are semantics-free. This gives rise to

disaligned, inconsistent and unrelated

information.

 entity update: an update on an entity does not

affect all the pages that semantically reference

it.

The semantic web is able to overcome these

difficulties and is gaining momentum day by day.

The rise of the semantic web has created a strong

demand by industry for people trained in semantic

technologies as well as ontologies. This demand is

estimated to quickly grow in the following years.

Paralleling this strong interest there is a flourishing

growth of courses in academia and professional

institutions; however, these courses are often single

courses within larger programs. The authors in

(Neuhaus, 2011) identified only one complete

academic program completely devoted to education

in applied ontology (a Master's program at the

University of Buffalo) and 21 programs that offer

ontology-centered topics.

Applied ontology and the semantic web are

relatively young disciplines and therefore there is

not yet a general consensus on questions related to

several aspects ranging from the curriculum to even

the terminology.

In the ACM/IEEE (ACM/IEEE, 2013) computer

science curriculum, for example, the knowledge

required for an ontologist is spread among courses

or topics within courses such as the elective

“Advanced Representation and Reasoning” in the

field of Intelligent Systems (IS), dealing with

ontology engineering and, in the Computational

Science field, the elective course entitled “Data,

Information and Knowledge”, dealing with

“Knowledge: ontologies, triple stores, semantic

networks, rules”, and so on. First, this paper will

briefly review teaching literature, curriculum

examples and didactic tools in the main areas of the

course: ontology design, ontology query languages,

ontology programming framework, mash-up and

semantic web applications. The design of an

introductory lab in semantic web technologies,

worth 3 ECTS, will also be presented.

Among the several collections of semantic web

educational resources to which the reader can refer

(Diederich, 2007), there exists a repository built by

the European Association for Semantic Web

Education (Diederich, 2006). In spite of this field

being relatively young, there are outstanding

experiences in teaching ontology engineering and

ontology design. In (Rewctor, 2004) the authors

report the results of a several-years-long experience

Giordano D. and Maiorana F..

Getting it Fast - An Information Systems Oriented Semantic Web Curriculum.

DOI: 10.5220/0004848300420048

In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 42-48

ISBN: 978-989-758-021-5

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

in teaching the OWL language and ontology design,

along with common errors and common design

patterns. Several ontologies have been used as

didactic examples to teach ontology design and

OWL, such as Pizza, Wines and Marsupials.

Recently, a didactic example in the biological

domain was proposed by (Schober, 2012), where the

authors propose an OWL ontology describing zoo

animals. The ontology, with respect to other large

biomedical ontologies, has the advantage of

massively reducing the number of classes and

overall complexity, allowing for fast memorization

and orientation, yet it uses all the major ontology

constructs, with related modelling challenges

together with some Ontology Design Pattern (ODP).

The proposed ontology uses a light version of Bitop

(Beißwanger, 2008) as a top level ontology.

Moreover, the ontology covers a domain that

involves common knowledge and therefore can be

easily understood by the majority of people,

especially in the life science domain. The ontology

was used in a complete curriculum for ontology

design described in (Boeker, 2012) where the

authors present a teaching experience of a one-week

workshop organized around 16 modules, addressing

4 main themes with an increasing level of design

complexity: basic principles, practical ontology

design, using top-level ontologies and ODP.

Various works address the development of

software tools used in training students, training and

teaching ontology query languages, and examples of

clear and didactically sound tutorials are also

available. A recent example of a software training

tool is described in (Gerber, 2010) where the authors

present a web-based SPARQL trainer that allows the

tutor to design a course along a set of concepts that

are to be tested. The tutor prepares a set of questions

and a dataset. Each question is provided with a

query solution against which the student solution is

compared. The comparison is based on the result

sets: the query solution and the student solution

result set must coincide both in the elements and in

their order.

For mash-up and semantic web applications, a

good tutorial can be found in (Della Valle, 2008)

where the author presents a semantic web

application that expects a music style as an input,

retrieves data from online music archives and event

databases, merges them by a bridge ontology and

allows the user to explore events related to artists

that practice the required style.

The aim of this paper is to give a detailed

overview of the curriculum taught in a Master's-level

class in Informatics Engineering in order to

“describe in more detail” how some of the

knowledge and skills indicated in (Neuhaus, 2011)

can be acquired. Our curriculum is “IT-oriented”

with a particular emphasis placed on the knowledge

and deployment of “IT systems involving many

components in addition to the ontology itself”

(Neuhaus, 2011). This type of curriculum fits well

with the knowledge and background of Informatics

Engineering and Computer Science students and

provides a means of broadening both data and

knowledge design capabilities, as well as system

integration and software development skills.

The paper is organized as follows: section 2

presents the teaching context and the contents;

section 3 describes the lessons learnt, highlighting

strengths and weaknesses of the approach used in

the course; section 4 reports conclusions and

describes further work.

2 TEACHING CONTEXT AND

CONTENTS

The teaching experience pertains to a 30-hour course

within the Informatics Engineering Master's degree

at the University of Catania. The course was taught

in the 2011 Fall term. The course was an elective

one, leading to a final certification. It was attended

by both first year and second year students.

Management of the class was not particularly easy

due to the students' varying levels of knowledge.

The course schedule was a three hour meeting per

week for ten weeks. The course was designed

around semantic technologies, with the aim of

demonstrating during the classes the main

technologies of the semantic web and the main tools

used to design and implement semantic applications.

As a university rule, the course was graded on a

pass/fail basis.

The students attending the course could get

acquainted with the following topics and questions:

1) Data transformation from a relational

database to RDF. After a review of the main tools,

Triplify (http://triplify.org/Overview) was chosen

due to its simplicity and its tight relationship with

database technologies and the PHP web

programming language, tools that are well-known to

all the students attending the course.

2) Querying triple stores/RDF files. The main

tool used was the SPARQL query language. The

Gruff browser was used both in a standalone version

and with the AllegroGraph server. Among the

different advantages of the Gruff browser, the graph

GettingitFast-AnInformationSystemsOrientedSemanticWebCurriculum

representation capabilities were considered a

definitive plus.

3) Relations and ontology design using RDFS

and OWL. Protégé was the tool chosen, due to its

flexibility, integration with reasoners, rich

availability of plugins and its free availability.

4) Linked Open Data: design and

implementation of a mash-up from linked data. The

Application Framework, Jena, was chosen.

The course also offered students possibilities to

familiarize themselves with the following core skills

and knowledge:

‐ Clarifying the purpose of a given ontology

‐ Judging what kinds of ontologies are useful for

a given problem

‐ Identifying, evaluating and using software tools

that support ontology development

‐ Using (reading and writing) different

representation languages

‐ Conducting ontological analyses

‐ Using a modern programming language

‐ Working in teams

The class activities were supported by an on-line

class site developed with the Moodle platform, used

for sharing lecture notes, projects and

communications and greatly simplified the

management of class activities.

The class schedule is reported in Table 1 with an

indication of the time required for each topic.

The table gives an overview of the curriculum

content that was organized into 4 main parts or

modules, with an increasing level of difficulty and

with a broader view of the covered topics. Each

module was further divided into 2 or 3 units. The

modularization of the curriculum allows for an

easier customization for other courses and classes

with different backgrounds.

The first module presents the Semantic Web in

comparison to the WWW and its main contributions.

The RDF language and RDF graph are presented

along with the major serialization languages: N-

triples, turtle and XML. Concepts like reification,

blank node and list and their impact in designing are

presented and applied to examples and exercises. An

introduction to the main ontological languages

RDFS, RDFS-Plus, OWL and their expressiveness is

undertaken.

The second unit deals with data transformation

and query techniques. The data transformation

technique was appreciated by the students and

served as a common thread between their strong

background on database and web programming and

the course content. After a review of the main tools,

such as D2RQ, Triplify was chosen due to its

Table 1: Class schedule.

Topic #H

Tools Readings Assignment

Part 1: Basic Principles

The Semantic

web

(Allemang,

2011) Ch. 1

RDF and

RDF graph.

Serialization

languages.

(Allemang,

2011) Ch. 2-3

Reification,

blank node,

list in RDF.

RDFS,

RDFS-Plus,

OWL

(Allemang,

2011) Ch. 3-4

Part 2: Data transformation and querying

Data

transformatio

n tools. Lab

session: use

of Triplify

Triplify

Allegro-

graph

Auer, 2009

Triplify a

Database

SPARQL and

SPARQL 1.1

Allegro-

graph/

Gruff

Ch. 5

(Allemang,

2011), Ch. 6

(Liyang,

2011),

(Bizer, 2009 )

Design a set

of queries

and evaluate

their

performance

Part 3: Ontology design

Ontology

engineering.

Use of

Protegè

Protegè/

OWLViz

Ch. 4 (Della

Valle, 2009),

Ch. 6, 7 8, 9,

11, 12

(Allemang,

2011) ch. 4,

5, 12 (Liyang,

2011)

Design an

ontology

The Jena

framework

Eclips/

Jena/

MYSQL

Ch. 12, 13,

14, 15

(Liyang,

2011),

(McBride,

2002),

(Carroll,

2004)

Develop an

application

that

integrates

data from

FOAF and

DBpedia

Part 4: Semantic web applications: mash-up, linked data and

visualization

Visualization

framework:

Exhibit

2 Exhibit Huynh, 2007

Visualize

events data

on google

maps

A semantic

web

application:

mash-up and

linked data

Combi-

nation of

tools

(Bizer, 2009

(b)) and

(Heath,

2011), Della

Valle, 2008)

Ch. (Della

Valle, 2009)

Develop a

mash-up

application

(optional)

Total 30

simplicity and its affinity with SQL and the PHP

development environment. This allowed students to

publish RDF triples from their own databases as

well as from data present in Bulletin Board, Content

Management systems and so on. The students

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

became acquainted with large RDF files and began

to design RDF graphs starting from an E/R schema.

The original work by (Auer, 2009) describing the

design of Triplify was used as a main reference. The

best works were presented by the students, which

benefited the entire class. The use of a shared pool

of projects, (Giordano, 2004) can be viable, even if

course is in its first edition, by using the projects

developed throughout the year, to facilitate the

students in overcoming difficulties, finding technical

details and solutions as well as to gain inspiration

from the work of their peers.

The subsequent step was the introduction of the

SPARQL language to query a RDF triple store. The

SPARQL language was selected since it resembles

the well-known database query language SQL. The

use of triples, graph patterns and other features

allowed the students to practice with the underlying

RDF graph and with serialization languages.

Moreover, the students were free to use OWL and

domain ontologies and to practice with large triple

stores such as DBpedia. For all these reasons the

introduction of SPARQL was considered a

beneficial step in this phase of the course. The

students were free to choose the RDF store where

they published their data. The suggested choice was

the AllegroGraph RDF store (http://www.franz.com/

agraph/allegrograph/) and the Gruff graph based

triple store browser (http://www.franz.com/

agraph/gruff/) due to the graph representation of the

results set. Other common choices were the Sesame

(http://www.openrdf.org/) and Joseky (http://

joseki.sourceforge.net/) RDF stores.

Chapter 5 of (Allemang, 2011) and chapter 6 of

(Liyang, 2011) were used as main references. All the

types of queries were covered. The unit started with

the select query form presenting triple and graph

patterns, query modifiers, optional patterns that

could be nested, filter conditions on different types

and regular expressions, union of graph patterns,

background and named graphs to query multiple

graphs. The other types of query forms, namely

construct and rules, describe and ask, were then

covered followed by the new SPARQL 1.1 features

such as aggregate functions, group by, sample and

bound, subqueries, negation, expressions with

SELECT, property paths, transitive queries,

federated queries and the SPARQL update 1.1

standard. In order to make the students aware of

performance issues in query formulation the

problem of benchmarking was presented through

studies such as (Bizer, 2009) and (Schmidt, 2009).

These works were also used as query examples.

Tools such as (http://ftp.heanet.ie/disk1/

download.sourceforge.net/pub/sourceforge/b/project

/bs/bsbmtools/bsbmtools/bsbmtools-0.2/) were used

by the students to profile queries and experiment

with different query formulations. The students

experimented with different datasets such as the one

suggested in (Bizer, 2009), FOAF RDF files, well-

known public domain ontologies such as the Wine

ontology and so on. Figure 1 shows an example of a

graph created with the Gruff browser and Table 2

shows a sample output of the bsbmtools available in

the Berlin SPARQL Benchmark.

The third part of the course was divided among

two related units. The first dealt with ontology

design principles. The main references were chapter

4 of (Della Valle, 2009) as well as chapters 6, 7 for

RDFS design and chapters 8, 9, 11, 12 from

(Allemang, 2011) for OWL-related design. Finally,

chapter 4 for RDFS design, chapter 5 for OWL

design and chapter 12 for tools and resources from

(Liyang, 2011) were also suggested to the students

for reference. Suggestions on how to present the

material are reported in the Lessons Learnt section.

As a working tool, Protégé was chosen

(http://protege.stanford.edu/) with the OWLViz

(http://www.co-ode.org/downloads/owlviz/) for

class visualization. The approach was to follow the

Protegè tutorial (Horridge, 2009) as well as sharing

suggestions such as those reported in (Rewctor,

2004) and (Noy, 2004), just to cite some examples.

An initial methodology and a practical example for

converting an E/R model into an OWL model was

presented, along with some ODP. Other literature

was suggested as further readings, i.e., (Poveda-

Villalón, 2010) and (Hammar, 2010). Two practical

ontology design exercises were also proposed, the

first one about the design of an educational portfolio

and the second one regarding the sharing of learning

objects. As a framework to develop semantic

applications and manage ontologies, the Jena

framework was chosen due to the large use of Java

technologies in both academia and the professional

world, as well as its free availability.

The Eclipse IDE and the MYSQL database to

store persistent models, were also suggested for data

manipulation in a database. As a reference for

practical applications, chapters 12, 13, 14 and 15 of

(Liyang, 2011) were suggested.

To present the Jena framework and its main

functionality and usage, the guidelines of

presentation indicated in (McBride, 2002) and

(Carroll, 2004) were followed. Following this in-

depth introduction on the architecture and main

features, practical examples on creating, using,

manipulating and navigating through RDF graph,

GettingitFast-AnInformationSystemsOrientedSemanticWebCurriculum

reading and storing to file as well as to database

were given with practical applications and working

code. Practical examples performing queries along

with using reasoners were also presented. The

tutorial used in the previous model was recreated

using, programmatically, the OWL API. Finally, a

complete example of a linked data application in the

style of (Liyang, 2011) was illustrated. The example

was used as a starting point for a linked data

application. The practical approach to linked data

was followed by a theoretical overview. The books

(Bizer, 2009 (b)) and (Heath, 2011) were used as

reference material and an overview of the main

concepts, design principles, techniques and recipes

for publishing and consuming linked data was

presented to the students.

(a)

(b)

Figure 1: An example of a SPARQL 1.1 query (a) with the

graph results visualized with the Gruff browser (b).

Finally, the fourth part completed the picture by

presenting visualization tools and strategies to

design and implement a mash-up semantic

application. Exhibit was selected as the visualization

tool due to is simplicity and flexibility in addition to

its free availability (Huynh, 2007) together with

tutorials and other learning material. In particular,

Exhibit 3.0 was presented to the students, together

with its use in a local environment.

In the original plan, all the technologies tools and

techniques had to converge in the presentation of the

Table 2: Result of the performance profile of the query on

the dataset described in (Schmidt, 2009) with 1.000 as

scale factor. The scale factor used is related to the number

of products.

Query: Find information of the Person whose label is

"reexhibit”

describe ?x

where {

?x rdfs:label "reexhibit". 

}

Output of the bsbmtools-0.2

Scale factor: 1000

Number of warmup runs: 20

Seed: 808080

Number of query mix runs (without warmups): 50 times

min/max Querymix runtime: 2.6075s / 3.1330s

Total runtime: 140.452 seconds

QMpH: 1281.58 query mixes per hour

CQET: 2.80904 seconds average runtime of query mix

CQET (geom.): 2.80503 seconds geometric mean runtime of

query mix

Metrics for Query: 1

Count: 50 times executed in whole run

AQET: 0.004793 seconds (arithmetic mean)

AQET(geom.): 0.004764 seconds (geometric mean)

QPS: 208.64 Queries per second

minQET/maxQET: 0.00429478s / 0.00822679s

Average result (Bytes): 1320.00

min/max result (Bytes): 1320 / 1320 

Number of timeouts: 0

final project. The aim was to present a mash-up

application that integrates different sources of data

in different formats, (e.g., relational, XML and

RDF) by designing a bridge ontology integrating all

these data in a unifying view. The reference material

was the tutorial (Della Valle, 2008) as well as

chapter 9 of (Della Valle, 2009) where the design

principles and main techniques and implementation

details are presented. This kind of application should

be considered as the concluding and unifying aspect

of the course, where all the principles and

technologies can be applied to build a working

semantic application.

The guiding factors for choosing the tools were:

the efficiency, the ease of use and its connection

with the database technologies for Triplify, the

visualization aids for Gruff, the free and open source

availability together with its large use in ontology

development for Jena, and the relative ease of use

and rich set of graphical features for Exhibit.

Overall there was a positive appreciation of the

course, particularly from the more motivated

students. Some of them expanded upon the course

content with a Master's thesis.

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

3 LESSONS LEARNT

Analyzing the teaching experience, it is possible to

make the following considerations. The initial

module on the semantic web, although an interesting

and necessary step that gives an overview of the

potentiality and breadth of the fields, can be reduced

to a minimum and allow the important concepts to

emerge from practical case studies such as querying

an RDF graph and ontology design.

The use of Triplify was a good ice breaking

activity allowing students to start from what they

already knew well, namely database design and web

programming, bringing them to fast production of

their RDF data and hence reading and navigating the

data produced.

The use of SPARQL for querying RDF store was

a definitive plus, serving as a bridge between

different concepts: it allows the student to master

serialization languages, ontology representation and

navigation, triple and graph pattern, query language,

and so on. The use of a benchmarking tool can be of

great benefit by allowing the students to place more

attention on query design and underlying execution

mechanisms. This type of knowledge can help in

obtaining a better understaning of both the querying

and, more broadly, of the ontology design processes.

The ontology design step is worth further

consideration. This can be done by allowing more

time for student projects and by sharing both their

projects and the peer reviews along with the trainer's

feedback and comments (Giordano, 2004). The use

of conversion from E/R to OWL is a good starting

point. Supplementing ODP by applying it to a real

case scenario could be of great benefit. The amount

of material suggested to the student should be

covered inside the development of a practical project

in order to avoid an overwhelming effect.

The programming side of the linked data and

mash-up applications can be the central aspect of

courses for Informatics Engineering and Computer

Science students. By leveraging their strong

background on software design and programming,

by scheduling the necessary time and granting a

valid number of credit units, there is the possibility

to develop, through group projects, interesting

examples of applications.

Some students reported the preference to begin

with ontology design and then view the querying

aspects. The best curriculum may then be designed

around a parallel development of query, ontology

design and programming framework to converge as

soon as possible with the design and implementation

of linked data and mash-up in a semantic web

application.

Due to the breadth and richness of the field along

with the cognitive load needed for mastering all the

topics and theoretical concepts as well as the set of

technical expertise, both as an IS engineer and as

designer and programmer, a course with more than

30 hours is required. In practice, due to an initial

underestimation of the overall workload, the

requirement for the final project was turned into an

optional one (only 10% of the students completed it)

and the course assessment was performed based on

three comprehensive exercises that demonstrated a

working knowledge of the techniques and tools

showcased during the course. With a course of 40 to

60 hours, more time could be devoted to practical

projects with a wider scope, to be developed both in

lab sessions, under the guidance of a tutor, and at

home, leaving the contents as described. Some

informal feedback collected by students who entered

the workforce after graduating and were assigned

responsibilities involving semantic web

technologies, commented that although the contents

were covered in a very condensed timespan, they

provided an effective groundwork.

4 CONCLUSIONS AND

FURTHER WORK

This paper has presented a curriculum, oriented

towards Information Technology, suited for a course

on semantic web and ontologies. The syllabus is

organized around 4 major modules allowing for easy

customization. The curriculum, by leveraging on the

database, software engineering and web

programming backgrounds of the participants, lays

the foundation for mastering semantic web

theoretical essentials, as well as the tools and

techniques necessary to develop applications that

exploit the full potential of this emerging field.

As further study a systematic analysis of the

mini-projects handed in by the students is planned,

to document common errors and pitfalls, for

example, in the query formulation with respect to

both the desired result and the execution efficiency,

to adjust accordingly further versions of this elective

course.

REFERENCES

Allemang, D., Hendler, J., 2011. Semantic web for the

working ontologist. Effective Modelingin RDFS and

GettingitFast-AnInformationSystemsOrientedSemanticWebCurriculum

OWL. Second Edition.(2011) Morgan Kaufmann.

Association for Computing Machinery, IEEE-Computer

Society, 2012. Draft Computer Science Curricula

2013.

Auer, S., S. Dietzold, Lehmann, J., Hellmann, S.,

Aumueller, D., 2009. Triplify - Light-weight Linked

Data publication from relational databases.

Proceedings of the World Wide Web Conference

(WWW) (2009), pp. 621-630.

Beißwanger, E et al., 2008. BioTop: An Upper Domain

Ontology for the Life Sciences - A Description of its

Current Structure, Contents, and Interfaces to OBO

Ontologies, In: Applied Ontology 3:4, pp. 205- 212.

Bizer, C.,Schultz, A., 2009. The Berlin SPARQL

benchmark. International Journal on Semantic Web

and Information Systems 5(2): pp. 1-24.

Bizer, C., Heath, T., Berners-Lee. T., (2009 b) "Linked

data-the story so far." International Journal on

Semantic Web and Information Systems (IJSWIS) 5.3

(2009): 1-22.

Boeker, M , Schober, D., Raufie, D.,Grewe, N., Rohl, J.,

Jansen, L., Schulz, S., 2012. Teaching good

biomedical ontology design. In Cornet R & Stevens R

(eds.), Proceedings of the 3rd International

Conference on Biomedical Ontology (ICBO 2012),

KR-MED Series, July 21-25, 2012, Graz, Austria.

URL http://ceur-ws.org/Vol-897/sessionJ-paper25.pdf.

Carroll, J. J., Dickinson, I., Dollin, C., Reynolds, D.,

Seaborne, A., Wilkinson, K., 2004. Jena:

implementing the semantic web recommendations. In

Proceedings of the 13th international World Wide

Web conference on Alternate track papers & posters,

pp. 74-83. ACM.

Della Valle, E., Celino, I., Cerizza, D., 2009. Semantic

Web: dai fondamenti alla realizzazione di

un’applicazione. Pearson – Addison Wesley.

Della Valle, E. 2008. RSWA 2008. Realizing a Semantic

Web Application. Tutorial presented at the 7th

International Semantic Web Conference (ISWC 2008),

Karlsruhe - Germany.

Diederich, J., M. Dzbor, Maynard, D., 2007. REASE—

The repository for learning units about the Semantic

Web. New Review of Hypermedia and Multimedia

13(2): pp. 211-237.

Diederich, J., Nejdl, W., Tolksdorf, R., 2006. EASE: The

European Association for SemanticWeb Education. In

Proceedings of the Semantic Web Education and

Training Workshop (SWET'06), co-located with the

First Asian Semantic Web Conference (ASWC 2006).

China: Beijing.

Gerber, D., Frommhold, M., Martin, M., Tramp, S., Auer,

S., 2010. Learning semantic web technologies with the

web-based SPARQLTrainer. In Proceedings of the 6th

International Conference on Semantic Systems pp. 26.

ACM.

Giordano, D., 2004. Shared values as anchors of a learning

community: A case study in information systems

design. Journal of Educational Media, 29, 3, 213-227.

Hammar, K., and Sandkuhl, K., 2010. The State of

Ontology Pattern Research - A Systematic Review of

ISWC, ESWC and ASWC 2005–2009. In: Blomqvist,

E., Chaudhri, V., Corcho, O., Presutti, V., Sandkuhl,

K. (eds.) Proceedings of the 2nd International

Workshop on Ontology Patterns - WOP2010. CEUR-

WS.org, Shanghai, China (2010).

Heath, T., Bizer, C., 2011.

Linked data: evolving the web

into a global data space. Morgan & Claypool

Synthesis Lectures.

Horridge, M., 2009. A practical guide to building OWL

Ontologies Using Protegé 4 and CO-ODE tools

Edition 1.3 available at: http://

owl.cs.manchester.ac.uk/tutorials/protegeowltutorial/

retrieved September 2013.

Huynh, D. F., Karger, D. R., Miller, R. C., 2007. Exhibit:

Lightweight Structured Data Publishing..

Liyang, Y. 2011. A Developer’s Guide to the Semantic

Web. Springer.

McBride, B., 2002. Jena: A Semantic Web Toolkit. IEEE

Internet Computing 6(6): pp. 55-58.

Neuhaus, F., Florescu, E., Galton, A., Gruninger, M.,

Guariono, N., Obrst, L., Sanchez, A., Vizedom, A.,

Yim, P., Smith, B., 2011. Creating the ontologist of

the future. Applied Ontologist, 6, pp. 91-98.

Noy, N., Mc Guinness, D. L., 2004. Ontology

development 101: A guide to creating your first

ontology. Stanford University.

Poveda-Villalón M., Suárez-Figueroa M.C., and Gómez-

Pérez A. 2010. A double classification of common

pitfalls in ontologies. Workshop on Ontology Quality

(OntoQual 2010), Co-located with EKAW 2010,

October 15, 2010, Lisbon, Portugal.

Rewctor, A., Drummond, N., Horridge, M., Rogers, J.,

2004. OWL pizzas: practical experience of teaching

OWL-DL: common errors & common patterns.

Knowledge in the age, Springer.

Schmidt, M., Hornung, T., Lausen, G., Pinkel, C., 2009.

SP2Bench: A SPARQL Performance Benchmark.

Schober, D., Grewe, N., Rohl, J., and Boeker, M. 2012.

ZooAnimals.owl: a didactically sound example-

ontology for teaching description logics in OWL 2.

OBML 2012 Workshop Proceedings.

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation