A SYSTEMIC METHODOLOGY FOR ONTOLOGY LEARNING

An Academic Case Study and Evaluation

Richard Gil, Leonardo Contreras

Dept. of Processes and Systems, Simón Bolívar University, Caracas-Baruta, Venezuela

María J. Martín-Bautista

Dept. of Computer Science and Artificial Intelligence, University of Granada, Granada, Spain

Keywords: Ontology Learning, Methodology, Systemic, Methodology Evaluation, Tools, Academic Domain.

Abstract: There is an important dispersion of technical and methodological resources to support the complete

Ontology Learning (OL) process from diverse knowledge sources. This fact makes the maintaining of the

structures of representation (ontologies) difficult. Therefore, the Knowledge-based Systems associated with

user’s domains may not fulfil the increasing knowledge requirement from the user. In this paper, we give a

possible solution for this problem. For this purpose, we propose a Systemic Methodology for OL (SMOL)

that unifies and simplifies to the users the whole process of OL from different knowledge sources

(ontologies, texts and databases). SMOL as methodology is evaluated under DESMET methods, in addition

with their application for an academic case study is also included.

1 INTRODUCTION

Reaching the knowledge from a semantic techno-

logical perspective has propitiated the development

of new technical methodologies and resources.

Through these product-tools obtained the society

should explore, discover, recover, store and update

knowledge associated to some specific domains

(Decker et al., 2000)(Nonaka, 1994).

The aim is not only to obtain system products to

support user requirements that may be ’devalued’

throughout time, but also to reach Knowledge-based

System (KBS) able to auto-learn and to make reco-

mmendations and learning actions related to

different user communities (Borges et al., 2008)

(Garruzzo et al., 2007).

However, this kind of systems is not so easy to

develop and maintaining (Abdullah et al., 2006).

This is due several aspects: first, the ontology

engineering methodological resources are in

maturation process yet; second, even though

technologies for handling knowledge based on

ontologies satisfy some user requirements, they

cannot guarantee a complete quality-driven and

user-oriented development during the ontology

engineering process; and finally, partial experiences

of ontology engineering developers, researchers and

users may be not incorporated as part of the tacit

knowledge (behaviour and skills) in the new metho-

dologies and technologies yet (Gómez-Pérez and

Manzano-Macho, 2005) (Haase et al., 2005).

Despite there are various definitions about OL,

we are in according to (Gómez-Pérez and Manzano-

Macho, 2005), where it is “the application of a set of

methods and techni-ques used for improving a

previous ontology with heterogeneous Knowledge

Sources (KSo), avoiding the complete Ontology

Development process”. These sources can be

ontologies previously developed, texts, database or

results of a process of ontology integration

(Maedche and Staab, 2001).

Precisely the widespread variety of mechanisms

and resources for OL make difficult the definition of

a standard methodology for OL. Consequently, in

(Gil, 2009) a new Systemic Methodology for OL

(SMOL) is conceived and proposed to overcome

some identified restrictions.

In this paper, we focus on the OL processes in

Section 2. The systemic methodology perspective is

showed in Section 3. SMOL methodology is descri-

206

Gil R., Contreras L. and J. Martín-Bautista M..

A SYSTEMIC METHODOLOGY FOR ONTOLOGY LEARNING - An Academic Case Study and Evaluation.

DOI: 10.5220/0003070602060212

In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2010), pages 206-212

ISBN: 978-989-8425-29-4

 2010 SCITEPRESS (Science and Technology Publications, Lda.)

bed in Section 4. A case study in the academic

domain applying SMOL is included in Section 5. A

general methodology evaluation is applied to SMOL

in Section 6, and finally, conclusions in Section 7.

2 ONTOLOGY LEARNING

There are several methodological alternatives in the

literature about OL. The one suggested in (Maedche

and Staab, 2001) includes some learning approaches

besides possible and recommended set of activities

associated with them. On the other hand, in (Gliozzo

et al., 2007), a different classification of the recom-

mended techniques into two groups are given. The

first group includes those approaches that allow get-

ting knowledge and retrieving information from

electronic texts. The second group, includes those

approaches that allow to 'gain knowledge' based on

previous structured knowledge and ontologies such

as dictionaries and thesaurus (Gómez-Pérez and

Manzano-Macho, 2005)(Gacitua et al., 2008).

OL approaches according with the KSo are

three-fold: a) OL from other ontologies developed

previously (Ehrig, 2007)(Noy and Musen, 2000)

(Euzenat et al., 2007) b) OL from documents (Buite-

laar and Cimiano, 2008) (Cimiano, 2006). And c)

OL from database schemes and their data-values

(Astrova et al., 2007)(Nyulas et al., 2007) (Cerbah,

2009).

2.1 Ontology Learning Resources

There are some definitions regarding Methodologi-

cal Resources (MR) that allows us to understand the

concepts associated to MR and to avoid confusions

that sometimes happen in technical literature. The

following definitions (Callaos, 1992) have been

considered: a) Techniques: subjective capabilities

(abilities or skills) to handle a tool properly. b) Me-

thods: a way of thinking or doing using a tool to

achieve an objective. c) Tools: objective capabilities

to use the resources properly to apply techniques.

And, d) Methodologies: a related set of methods,

techniques and tools which could be used for

reaching objectives.

2.2 Ontology Learning Problematic

Although important technical advances in MR in the

OL field, according to each KSo, have demonstrated

the main OL strengths and opportunities, authors

recently have reported high dispersion and little

integration among those MR producing OL results

from the same KSo.

Therefore, to synthesize the general OL

problems a situational technical analysis, which is

known as SWOT (Strengths, Weaknesses,

Opportunities and Threats), has been used (Hill and

Westbrook, 1997). This technique simplifies the OL

understanding from two broad perspectives. First, it

addresses the knowledge development and

reconstruction as an OL process and, secondly, by

studying it in terms of the resulting semantic (Noy

and Klein, 2004).

In agreement with (Gómez-Pérez and Manzano-

Macho, 2005) and (Shamsfard and Abdollahzadeh,

2003) some conclusions associated with those

studies about OL methodologies can be summarized.

Regarding OL Methods: a) There is not an esta-

blished standard. b) The methods are not usually

combined, and c) Many methods are not associated

with specific tools. With regard to OL Tools: a) All

of them help to extract knowledge; b) A small group

of them allow the retrieval of a complete taxonomy,

and c) Only some tools support specific OL

methods.

It is possible to infer also, that OL methodology-

cal options do not exist as a complete integral,

unified and dynamic way to face the OL problems

for knowledge recovery from heterogeneous KSo.

3 SYSTEMIC PERSPECTIVE

Methodological options used to get designs and

knowledge product development (systems, models

or ontologies) are associated to strategies and pro-

cesses structured in some way. Many approaches

closer to Software Engineering reflect the efforts

dedicated in this direction (Sommerville, 2006).

More specific methodological approaches orient-ted

on one hand to Software Development (Press-man,

2006) and on the other hand, to Knowledge

Engineering (Gómez-Pérez et al., 2004) (Buitelaar

and Cimiano, 2008)(De-Nicola et al., 2009) the

previous methodological options have arisen.

The proposed methodological perspective tries to

conciliate the system development total quality

paradigms with user-centered services to attend their

demanded requirements. This conciliation is suppor-

ted by systemic methodologies instead of systematic

ones (Callaos and Callaos,2003): a) Systematic

methodologies are oriented to the efficiency, with a

predetermined behaviour, strict and closed. (e.g.

Structured Life Cycle) and b) Systemic methodolo-

gies are oriented to the effectiveness, with a non-

A SYSTEMIC METHODOLOGY FOR ONTOLOGY LEARNING - An Academic Case Study and Evaluation

207

Figure 1: Systemic Methodology for Ontology Learning applied for an academic case study.

predetermined behaviour, flexible and open. (e.g.

Agile Process).

Applying a methodology to an environment (an

organization or system) is an evolutionary maturing

process to achieve results with approaches such as

the Action-Research (Baskerville, 1999) and the

Action-Learning (Dilworth, 1998). The action (Ac-

tion-Design) in both cases allow to support: first,

researching to discover; second, learning to unders-

tand and to experiment; and third, system design and

synthesis to generate new ideas and to solve specific

problems to satisfy certain requirements.

Indeed, the product (ontologies) and the process

(methodologies) must be developed in a trade-off

between efficient and effective action-design.

4 DESCRIPTION OF SMOL

The lack of integrated methodologies covering the

whole process of OL leads us to propose and

experiment with new methodological options. This

new systemic methodological proposal must be

flexible, iterative, incremental and adaptable to

normal users, experts and knowledge engineers

using some MR previously developed according the

quality approach cited (Callaos and Callaos, 2006).

Users of SMOL to combine MR for diverse KSo

in a proper way, considering the existence of a

domain ontology already elaborated for KBS which

could be improved through updating/enrichment OL

processes (Haase et al., 2005)(Noy and Klein, 2004).

For this methodology design, we select a frame-

work for knowledge retrieval of (Yao et al., 2007).

4.1 SMOL Phase-flow Description

The phase-flow of SMOL is proposed (some in

Figure 1), emphasizing the MR recommended to be

used in each specific phase. The activities related to

each original phase of SMOL are explained as

follow: I. Methodology strategy selection. The

complexity of the domain is evaluated based on the

availability of information/knowledge useful about

the domain (Zhou, 2007). The methodology strategy

is drafted/selected using an appropriated arrange-

ment of MR for each KSo relative to. II. Knowledge

discovery. The MR from different knowledge -sour-

ces and -repositories are combined. III. Query requi-

rements. Different queries are formulated to the KSo

available by browsers or other kind of applications.

IV. Knowledge selection. A selection of the retrie-

ved information from the formulated queries to the

sources and repositories is performed. V. Know-

ledge structures construction. Different structures

such as ontologies and contexts can be built interact-

tively with users’ advisory by ontology alignment,

machine learning techniques, etc. VI. Knowledge

exploring and searching. The knowledge structures

are explored, verified and validated and the search

can be refined. VII. Knowledge structures reorgani-

zation. Processes such as grouping of instances,

ontology population and other similar activities are

performed in this phase. And, VIII. Knowledge-

based System configuration. Users set-up the main

modules of the KBS that have ontologies updated

and associated with the users’ domain.

Other five activities were developed for SMOL

drafting: 1) Methodology strategic selection phase is

designed considering that the user of SMOL may

adjust MR according to the information available in

the KSo about the domain-complexity. 2) Knowled-

I- Methodology

trategy selection

(bottom-up

)

II- Knowledge

Discovery

(Scholar-Google)

III- Query

requirements

(Classes taxonomy)

IV-Knowledge

selection

(By Rapid-I)

V- Knowledge struc-

tures construction

(GATE-Ontology)

VII-Knowledge struc

ture reorganization

(Integration DEA+Text)

VI- Knowledge ex-

ploring & searching

(+Text)

Knowledge

structure updated

Satisfied requirements

Strategy selected

Texts

Structured

knowledge

Knowled

e sources

User/Task Profiles

Users

Ontology-DEA+Texts

Phase

Decision point

Phases Flow

Data flow

Storage-DB

KEOD 2010 - International Conference on Knowledge Engineering and Ontology Development

208

ge Sources are configured as storage component

(DB) with the purpose of knowledge reusing. 3)

User/Task profiles are configured as storage compo-

nents to queries-operations registration (log) with

the purpose of reusing MR and recommending tasks.

4) Decision points have been included for user cycli-

cal-quality-check purpose. Some of them are shown

–as rhombus- in the Figure 1. And, 5) A methodolo-

gical Phase-flow activities description is detailed as

input, output, methods and tools recommended

according with each KSo and strategy selected.

5 ACADEMIC CASE STUDY

The objective of this study in the academic domain

is to retrieve and to add new knowledge into an

ontology named Ontology-DEA previously develo-

ped for a Decision-Support System of a University

of Venezuela (Ramos and Gil, 2007). In this experi-

mental case, it was improved by users the ontology-

DEA (incremental/iterative) with knowledge extrac-

ted from a corpus of texts. SMOL application for

this case is shown in Figure 1 (Gil et al., 2009).

A bottom-up learning strategy (Phase I) was

draf-ted and selected considering the following key

acti-vities: a) Finding and selecting a set of texts

from Internet with experts-users advisory. b)

Identifying from the corpus, some relevant keywords

by agent for ontology updating. And, c) applying the

OL from texts via text annotations and ontology

population.

Texts selection with user’s participation is

carried out in these Phases (II & III ). Users

recovered texts for the corpus, through Google

Scholar. From an initial set of 1000 retrieved texts, a

final set of 480 texts were selected using a file-

length base.

The learning agent developed in RAPID-I with

the plug-in WVtool was used to classify texts by

their relevant keywords, so they could be added to

the corpus for future updating (Phases III & IV). The

used technique is “text clustering” with the TF-IDF

term weighting scheme. Moreover, different pro-

cessses of tokenization, stop-word removing and

stemming were performed. Keywords found by the

agent were: “accredit, style, programming, distance,

institute, program, online, faculties, course and

student”. Those Keywords selected by the agent

were inputs to the next process in GATE via Onto

Gazetteers (Phase VI). The central purpose was to

identify representative terms and concepts in the

texts of the corpus besides corresponding Gazette-

ers’ annotation standard (e.g. dates or places).

An ontology graphical tool option for ontology

management was used in GATE to display annota-

tions to the users and to help them to support ontolo-

gy updating (evolution) from texts (Phases V &VII).

Other SMOL applications for the same case

study using other KSo have been reported: a) OL by

comparing to domain ontology located and

recovered from the Internet (Gil et al., 2008). And,

b) relevant knowledge about profiles of professor’s

subdomain from a relational database (RDB) of

another University was obtained (Gil et al., 2010).

6 SMOL EVALUATION

There are not so many alternatives for methodology

evaluations applied to the Ontology Development

field. One of the most referred in the Software Engi-

neering area is DESMET (Kitchenham, 1996).

We have used a combination of these DESMET

methods: a) Screening: A feature-based evaluation

done by a single individual who not only determines

the features to be assessed and their rating scale but

also does the assessment. For initial screening, the

evaluations are usually based on literature describing

the software method/tools rather than the actual use

of the methods/tools. b) Experiment: A feature-

based evaluation done by a group of potential users

who are expected to try out the methods/tools on

typical tasks before making their evaluations. And,

c) Case study: A feature-based evaluation performed

by someone who has used the method/tool on a real

project.

Those methods are recommended by DESMET

to be used when: a) Large number of methods/tools

to assess. b) Short timescales for evaluation exercise.

c) Benefits difficult to quantify. d) Benefits

observable on a single project. e) Stable

development procedu-res. f) Relatively small

learning time. And, g) Tool-/method user population

very varied and limited.

6.1 Qualitative Screening

To apply the Qualitative screening of the DESMET

evaluation, we have followed the two-fold: First, we

have performed below an interesting evaluation

approach applying usability/suitability criteria

assessment to evaluate by comparing our proposal

(Dahlem and Hahn, 2009).

On the other hand, we have developed a short

comparison of SMOL among two similar OL

methodologies published recently by (Simperl et al.,

2008) and (Novacek et al., 2007).

A SYSTEMIC METHODOLOGY FOR ONTOLOGY LEARNING - An Academic Case Study and Evaluation

209

6.1.1 Screening through Usability Criteria

Dahlem’s usability and suitability evaluation pro-

posal (Dahlem and Hahn, 2009) has considered the

thirteenth methodological criteria suggested and

applied. Those are: Adequate terminology (C1);

Structure (C2); Descriptiveness (C3); Transparency

(C4); Error avoidance (C5); Robustness (C6);

Lookahead (C7); Consistency (C8); Hiding forma-

lity (C9); Expressiveness (C10); Conceptualization

flexibility (C11). Ontology assumptions (C12); and,

Tool support (C13). These criteria are combined

originally in an upper level under the following

terms: I) Learnable. II) Efficiency. III) Memo-

rability. IV) Error Handling. And, V) Satisfaction.

First, we assessed the SMOL methodology cha-

racteristics with the usability criteria cited. As result,

SMOL has up to nine of thirteen representative

criteria for methodology usability according to

Dahlem’s proposal. According to the total usability

evaluation criteria (uniform presence considered),

Efficiency and Satisfaction for SMOL, the value is

high (0,85 & 0,75). As for the Learnable, Memora-

bility and Error-handling criteria about SMOL are

medium (0,5). Indeed, these methodology evaluation

results about SMOL show comparative feature-

/capabilities among other equivalent methodologies.

6.1.2 Screening through Comparison

Comparing the SMOL methodology among the

proposal of Simperl et al. (Simperl et al., 2008) and

DINO (Novacek et al., 2007), there are some aspects

to point out. Mainly, SMOL has more elaborated

methodological options to support OL processes

from diverse KSo. The main options are: 1) SMOL

considers explicitly the assessment of the domain-

complexity characteristics for strategy selection.

And, 2) the OL strategy selection is based on an

approach of learning to start not always from texts,

but from other KSo such as databases and

ontologies, inclusive at beginning of the process.

Some details are omitted due paper-pages limit.

6.2 Qualitative Experiment And Case

The main way to test the SMOL functionality is

based on the case study, because we can check the

user validation and experiment with related me-

thods/tools. Some users were trained/familiarized

with some OL methods/tools used (e.g. Protégé,

Prompt-CogZ, Racer-Pro, GATE).

For each KSo, an evaluation strategy has been

designed considering: context (goals-constrains),

planning and design, preparation, execution, data

analysis, dissemination and decision-making.

An interview-questioner was given to the users

(up to 6) during the OL cycle according to each

KSo. A feature-based analysis was applied to those

results associated with those MR used. Particularly,

we asked them about Tools-functionality and Input-

/Output related to the OL methods/techniques

learned/applied.

The case study evaluation revealed the user satis-

faction about the SMOL methodology flexibility,

due to the capability of the MR integration in the

systemic component. A minor issue very interesting

for the user is the diversification of strategies to

reach knowledge aggregation from different KSo.

7 CONCLUSIONS

There is a lack of integrated methodologies in the

OL process, whatever the sources considered: onto-

logies, texts and database. A Systemic Methodology

for OL named SMOL has been designed considering

pros and drawbacks of the previous OL methodolgi-

cal proposals but including MR for diverse KSo. The

result is an integral, flexible, open, interactive and

iterative methodology user-oriented.

The SMOL methodology has been applied and

checked in an academic case study for different

KSo. Particularly, the OL from texts has been

detailed in this work.

The ontology updated by user’s participation

help us for SMOL validation. The SMOL

methodology evaluation as well as the preliminary

result obtained for this case study, reveal the

feasibility of SMOL as an instance of new

methodological perspectives for OL from texts, as a

way to update ontologies asso-ciated with KBSs of

the users’ domain.

In future works, promissory results could be ob-

tained with other SMOL cases applications combi-

ning incrementally some different KSo. Likewise,

other experimental and specific evaluations have

been performed to increase the SMOL background.

REFERENCES

Abdullah, M., Kimble, C., Benest, I., and Paige, R. (2006).

Knowledge-based systems: a re-evaluation. Journal of

Knowledge Management, 10 Nro 3:127–142.

Astrova, I., Korda, N., and Kalja, A. (2007). Rule-based

transformation of sql relational databases to owl

KEOD 2010 - International Conference on Knowledge Engineering and Ontology Development

210

ontologies. Proceedings of the 2nd International

Conference on Metadata & Semantics Research.

Baskerville, R. (1999). Investigating information systems

with action research. Comm.AIS, v:2: Art 19.

Borges, A., Corniell, M., Gil, R., Contreras, L., and

Borges, R. (2008). Towards a study opportunities

recommender system in ontological principles-based

on semantic web environment. In The 4th

WSEAS/IASME. (EDUTE’08), ACM vol 8, no 2.

Buitelaar, P. and Cimiano, P. (2008). Ontology Learning

and Population: Briging the Gap Between Text And

Knowledge. IOS Press, Netherland.

Callaos, N. (1992). A systemic system methodology. In

International Conference on System Research

Informatic and Cybernetics, Baden-Baden, Germany.

Callaos, N. and Callaos, B. (2003). Toward a practical

general system methodological theory. Journal of

Systemics, Cybernetics and Informatics, 1:114–120.

Callaos, N. and Callaos, B. (2006). Designing with a

system total quality. In on Information System

Analysis, and Synthesis, ISAS’06, USA, p 15–23.

Cerbah, F. (2009). RDBToOnto User Guide, Version 1.2

Beta From relational Database to Fine-Tuned

Populated Ontologies. http://www.tao-project.eu/

Cimiano, P. (2006). Ontology Learning and Population

from Text: Algorithms, Evaluation and Applications.

Springer-Verlag New York, LLC.

Dahlem, N. and Hahn, A. (2009). User-friendly ontology

creation methodologies-a survey. 15th Amer. Conf.

On Information Systems, California-USA, pages 1–9.

De-Nicola, A., Missikoff, M., and Navigli, R. (2009). A

software engineering approach to ontology building.

Information Systems, Elsevier, 34:258–275.

Decker, S., Melnik, S., Van-Harmelen, F., Klein, D.,

Fensel, F., Broekstra, M., Erdmann, J., and Horrocks,

M. (2000). Knowledge networking the semantic web:

The roles of xml and rdf. IEEE Internet Computing.

Dilworth, R. (1998). Action learning in a nutshell.

Performance Improvement Quarterly, Vol 11(1).

Ehrig, M. (2007). Ontology Alignment: Biding the

Semantic Gap. Book, Springer-Verlag.

Euzenat, J., Mocan, A., and Sharffe, F. (2007). Ontology

Management, Ontology Alignments: An Ontology

Management Perspective. Book, Springer.

Gacitua, R., Sawyer, P., and Rayson, P. (2008). A flexible

framework to experiment with ontology learning

techniques. Know.-Based Syst., 21(3):192-199.

Garruzzo, S., Rosaci, D., and Sarné, G. (2007). Mars: An

agent-based recommender system for the semantic

web. LNCS, 4531:181–194.

Gil, R. (2009). New systemic methodology framework for

ontology learning (in spanish). Master’s thesis,

Dpto.Computer Science, Granada University, Spain.

Gil, R., A.Borges, Ramos, L., and Contreras, L. (2008).

Ontologies integration for university institutions:

Approach to an alignment evaluation. In Proc. 19th

Australian Conference on Software Engineering

ASWEC 2008, pages 570–578.

Gil, R., Borges, A., Contreras, L., and Martín-Bautista, M.

(2009). Improving ontologies through ontology

learning: a university case. In CSIE’09, IEEE

Computer Society, March-April, L.A.-USA.

Gil, R., Martín-Bautista, M., and Contreras, L. (2010).

Applying an ontology learning methodology to a

relational database: University case study. IEEE-

ICSC-2010, USA, Sep 22-24 (Accepted).

Gliozzo, A., C. Caracciolo, M. D-Aquin, M. S., Peter, W.,

Voelker, J., Dzbor, M., Mota, E., Gomez-Perez, A.,

Haase, P., Waterfield, W., Contreras, J., Grobelink,

M., Euzenat, J., Cunning, H., Staab, S., Gangemi, A.,

Angele, J., Iglesias, M., Lobo, T., and Lopez, A.

(2007). Results from experiments in ontology learning

including evaluation and recommendation. Technical

Report, http://www.neon-project.org/

Gómez-Pérez, A., Fernando-López, M., and Corcho, O.

(2004). Ontology Engineering. Book, Springer-Verlag,

London- UK.

Gómez-Pérez, A. and Manzano-Macho, D. (2005). An

overview of methods and tools for ontology learning

from text. Knowledge Engineer. Rev., 19:187–212.

Haase, P., Volker, J., and Sure, Y. (2005). Management of

dynamic knowledge. Journal of Knowledge Mana-

gement, 9:97–107.

Hill, T. and Westbrook, R. (1997). Swot analysis: It’s time

for a product recall. Long Range Planning, 30:46–52.

Kitchenham, B. (1996). Evaluating software engineering

methods and tool. part 3: Selecting an appropriate

evaluation method. ACM SIGSOFT Software

Engineering Notes, 21:9-12.

Maedche, A. and Staab, S. (2001). Ontology learning for

the semantic web. Intel. Systems, IEEE, 16:2:72-79.

Nonaka, I. (1994). A dynamic theory of organizational

knowledge creation. Organization Science, 5(1):14-37.

Novacek, V., Laera, L., and S.Handschuh (2007).

Semiautomatic integration of learned ontologies into a

collaborative framework. In Proceedings of

IWOD/ESWC, and ESWC 2007.

Noy, N. and Klein, M. (2004). Ontology evolution: Not the

same as schema evolution. Knowledge and

Information Systems, 6:428–440.

Noy, N. and Musen, M. (2000). Prompt: Algorithm and

tool for automated ontology merging and alignment.

Proceeding of National Conference On Artificial

Intelligence.

Nyulas, C., O’Connor, M., and Tu, S. (2007). Datamaster

- a plug-in for importing schemas and data from

relational databases into protege. 10th Intl. Protege

Conference, Budapest (2007).

Pressman, R. (2006). Software Engineering: A Practioners

Approach. Sixth Edition McGraw-Hill, New York

2006.

Ramos, L. and Gil, R. (2007). Propuesta de sistema de

información para apoyar la gestión de la educación a

distancia. CISCI’07, July. Orlando-USA.

Shamsfard, N. and Abdollahzadeh, A. (2003). The state of

the art in ontology learning: a framework for

comparison. Know. Enginer. Rev, 18 (4):293-316.

Simperl, E., Tempich, C., and Vrandecic, D. (2008).

Methodology for Ontology Learning. Chapter of Book

A SYSTEMIC METHODOLOGY FOR ONTOLOGY LEARNING - An Academic Case Study and Evaluation

211

Ontology Learning & Population. IOS Press, Buitelaar

& Cimiano Eds.

Sommerville, I. (2006). Software Engineering. Pearson

Education.

Yao, Y., Zeng, Y., Zhong, N., and Huang, X. (2007).

Knowledge retrieval. In Proceedings IEEE/WIC/ACM

International Conference on Web Intelligence.

Zhou, L. (2007). Ontology learning: state of the art and

open issues. Information Technology and Manage-

ment, 8(3):241–252.

KEOD 2010 - International Conference on Knowledge Engineering and Ontology Development

212