Semantic Representation of Information by Ontological Networks to

Improve Knowledge Management in Higher Education

Roberto A. Garc

ıa-V

elez

, Jorge Andr

es Gal

an-Mena

, Ahmed Dahroug

Vladimir E. Robles-Bykbaev

and Mart

ın L

opez-Nores

GI-IATa, C

atedra UNESCO Tecnolog

ıas de apoyo para la Inclusi

on Educativa, Universidad Polit

ecnica Salesiana,

Calle Vieja 12-30 y Elia Liut, Cuenca, Ecuador

Arab Academy for Science, Technology and Maritime Transport, Cairo, Egypt

AtlanTTic Research Center for Telecommunication Technologies, University of Vigo, Vigo, Spain

Keywords:

Ontologies, Higher Education, Knowledge Management, Semantics.

Abstract:

Institutions of Higher Education (IHE) seek to respond to new ways of conceiving and projecting higher edu-

cation embodied in the proﬁle of a professional. Faced with this challenge, educational models focus mainly

on the theoretical-practical references of critical training, constructivism and collaborative learning. Although

there are several knowledge management solutions in the ﬁeld of higher education, which have managed to

formalize the organizational structure of academic institutions in ontologies, so far none of these proposals

divides and makes explicit the development that the actors of the ecosystem (students and educators) take over

time. Our proposal is to construct a semantic representation of the academic ecosystem by implementing an

ontological network that allows managing the knowledge generated in a more efﬁcient way. This architecture

is intended to be used as an instrument to support the academic body and for the centralization of information.

To achieve our objective, we carry out a reengineering process of relevant institutional documents, such as the

academic record of the students, in each of their facets of learning. We have interviewed specialists in the area

and reuse academic domain ontologies to form a consistent knowledge base.

1 INTRODUCTION

Higher education institutions periodically generate

large volumes of data associated with the academic

ﬁeld (Mora-Arciniegas et al., 2017). Unfortunately,

the information that revolves around the university

academic environment is usually distributed among

multiple sources of subsystems that use different da-

tabases or digital repositories (Mora-Arciniegas et al.,

2017; Aminah et al., 2017). Having a diversity of un-

structured formats from different sources of informa-

tion generates a restriction of interoperability among

them. This particularity restricts the ability to efﬁ-

ciently manage and exploit the full potential of this

information in the form of knowledge, which could be

used in a timely and relevant manner by the different

actors of the academic ecosystem. Although there are

several knowledge management systems in the ﬁeld

of higher education focused on solving this difﬁculty

(among them, transactional systems that allow associ-

ating information between different subsystems), the

information retrieved does not have semantic relati-

onships that would allow subsystems to identify even

whether two pieces of information relate to the same

topic or entity.

Based on this background, in order to help univer-

sities ensure an efﬁcient management, sharing, search

and reuse of information, we have proposed the crea-

tion of an ontological network that allows linking the

information generated and take it to a higher level,

making sure the right information gets to the right pe-

ople at the right time to make the right decisions. In

this paper, we describe an experiment in which these

ontologies supported knowledge management within

the Universidad Polit

ecnica Salesiana (UPS), located

in the city of Cuenca, Ecuador. From our ﬁndings, we

analyze the strengths of knowledge administration in

higher education supported by semantic technologies.

We also suggest some avenues to support this activity

within higher education institutions.

The idea of using semantic technology is to enable

universities to improve the mechanisms for searching

for information and knowledge. This is particularly

necessary for academic institutions of higher educa-

García-Vélez, R., Galán-Mena, J., Dahroug, A., Robles-Bykbaev, V. and López-Nores, M.

Semantic Representation of Information by Ontological Networks to Improve Knowledge Management in Higher Education.

DOI: 10.5220/0006920801610168

In Proceedings of the 10th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2018) - Volume 3: KMIS, pages 161-168

ISBN: 978-989-758-330-8

161

tion, where sharing, reusing and inferring new know-

ledge plays a central role.

In Section 2, some related works developed in

the ﬁeld of higher education institutions are descri-

bed. In Section 3, the development of the proposal

is detailed indicating the methodology used (which

implies the speciﬁcation of requirements for the on-

tology network), the reuse of resources, the reengi-

neering carried out to reuse the ontological resources

and the conceptualization of the ontological network

proposal based on the UPS datasets. Continuing, in

Section 4 we present the results of the experimenta-

tion, and Section 5 contains the conclusions of our

work.

2 RELATED WORK

There is a broad ﬁeld of work related to knowledge

management in the academic domain using ontolo-

gies. Several authors have used them as a means to

represent different perspectives of the academic dom-

ain. For example, the authors of (Laouﬁ et al., 2011)

proposed an approach to establish an organizational

Memory of the University System (MUS). The MUS

is not only on pedagogical knowledge (topics, sub-

jects, courses, etc.) but also works with other know-

ledge used in the context of a university system, whet-

her administrative, educational or related to the intel-

lectual capital of the university. For the construction

of the ontology, the actors are trained to explain their

knowledge, promoting in this way a collaborative cul-

ture of creation and indexation of knowledge. The

captured knowledge provides a resource dedicated to

the representation of ontological knowledge, which

facilitates the search and navigation between concepts

related to the academic system. The architecture of

the system is able to integrate different ontologies,

and enables different types of reasoning and intelli-

gent information retrieval.

A similar approach was done by the Universidad

ecnica Particular de Loja (UTPL) research univer-

sity (Mora-Arciniegas et al., 2017). They propose

the use of semantic technologies to represent acade-

mic knowledge through ontologies, creating a set of

comprehensible and interoperable data. To fulﬁll their

objective, they developed an ontological network cal-

led Linked Academic Data (LDA) for the representa-

tion of the organizational structure and teaching plan-

ning. LDA was comprised of ontological resources

(FOAF, AIISO, VCARD, ORG, DCTERMS, VIVO),

non-ontological resources that required reengineering

to transform them into accessible resources, and the

creation of a new vocabulary.

A more recent proposal can be found at (Melgar

and Quilca, 2016), consisting in an architecture for

organizational memory systems in higher education

institutions. The proposed architecture for organiza-

tional memory is based on CESM model and Com-

monKADS methodology, concerned with the kno-

wledge representation for semantic search of docu-

ments. This approach allows adding semantic content

to the documents and to the information, enabling the

construction of knowledge bases. The semantic con-

tent added to the documents allows the retrieval of

information based on inference, unlike the relational

database that is based only on the content of a ﬁeld.

To achieve this, they add semantic annotations to do-

cuments and link them to deﬁned ontologies.

Finally, the authors of (Gasmi and Bouras, 2018)

proposed a system that allows to improve vocational

education, conducting an analysis of the gap between

education and industry. To this aum, they used a sy-

stem of semantic concordance based on ontological

models that allows to analyze these two domains. In

particular, the educational and industry proﬁles are re-

presented as proﬁles of O*NET competency frame-

works, which allows to analyze the gap according to

the structure of higher education and the peripherals

required by the industry in the social context.

3 DEVELOPMENT

In order to unify the scattered information in several

database repositories that occurs around the different

learning processes of higher education institutions,

we propose the construction of a new formal model

that captures the corresponding knowledge. We chose

the NeOn framework (Su

arez-Figueroa et al., 2012b)

as the basic was chosen for our implementation, be-

cause it offers features that allow to manage the life-

cycle of the ontological network very easily.

3.1 Elicitation of Requirements

In order to identify the purpose of creating the formal

model, who the end users would be and the features

it would enable, we went through a process of speci-

fying ontological requirements (Su

arez-Figueroa and

omez-P

erez, 2012). To fulﬁll the different tasks

speciﬁed by the NeOn methodology, a group of spe-

cialists was formed, composed of practitioners and

specialists from the academic side of higher educa-

tion institutions, who issued their criteria according

to their role in each task. Table 1 presents the group

of most frequent terms obtained from an interview for

KMIS 2018 - 10th International Conference on Knowledge Management and Information Sharing

162

Figure 1: Modular scheme of the ontological network of the academic ecosystem.

the identiﬁcation of functional requirements, which

were already grouped and validated.

Table 1: Group of most frequent terms obtained.

Code Item Competence Questions

TG87 Micro curriculum PC18, PC26

TG100 Class period PC11

TG114 Task PC7, PC22

TG41 Teacher PC1, PC3, PC6

3.2 Formalization of the Scheme

Starting from the requirements elicitation process,

we sought existing resources to reuse and reengi-

neer (Su

arez-Figueroa et al., 2012a), ending up with

the following selection:

• Friend of a Friend (FOAF) (Kalemi and Martiri,

2011) which describes a basic social network of

people and their relationships;

• VCARD (Iannella and McKinney, 2014) is an on-

tology that describes information of people and

organizations

• VIVO has a focus for a community of research.

• Basic Formal Ontology (BFO) (Arp et al., 2015)

is a top-level ontology that helps the integration of

multiple ontology in a single formal framework.

In the construction of our ontological network it

was decided to modularize the classes, both those

built from the reengineering of non-ontological do-

cuments and the selected ontologies to be reused.

Figure 1 shows the scheme of our network, where

we have reused VCARD, FOAF, BFO and VIVO

ontologies by means of different links and selection

of components according to the necessary domain

through a process of alignment and modularization of

ontologies. Our academic ecosystem ontology (AEO)

is separated into 5 modules: Micro Curricular (MC),

Roles (R), Academic Metrics (AM), Metrics (M) and

Occurrent (O). These modules participate as links

between the reused ontologies, which have different

dependencies and cover speciﬁc domains within the

same context.

The main classes deﬁned in our ontology are the

following:

• Person: It is a representation of a person, reused

from the FOAF ontology.

• Evaluated Role: Inherent role to a person that

fulﬁlls the function of being evaluated within a

process.

• Subject: Representation of a subject that is

taught in a course within a study program.

• Career: Consensus of the name that is given to

a study program that leads to the granting of a de-

gree to practice a profession.

• Curriculum: Organization of the subjects and

requirements of a career that determine the pro-

Semantic Representation of Information by Ontological Networks to Improve Knowledge Management in Higher Education

163

fessional proﬁle of the student.

• Tuition: Representation of the document of re-

cord of notes of a student enrolled in a subject be-

longing to a curriculum.

• Tuition Status: status of a license plate, it can

be: approved, failed or canceled.

• Identity Card: Identity document that belongs

only to one person.

Each one of the modules of the ontological net-

work belong to a speciﬁc context, allowing us to bet-

ter manage the different components imported into

our network. For the construction of the modules, the

Prot

tool was used as shown in Figure 2, where

each one of the ontologies is developed separately ta-

king into account the alignment of the components

through links that are classes of other modules. To ve-

rify the operation of the network as a ﬁrst step, all the

modules were imported into the same project. Finally,

we ran the HermiT

reasoner to verify that there were

no inconsistencies in the axioms.

Figure 2: Prot

e capture of the classes of the ontological

network modules of the academic ecosystem.

3.3 Information Mapping

The construction of the ontological network implies

both the formalization of the RDF (Klyne and Carroll,

2006) scheme. Starting from this referential semantic

framework we start with the population of the onto-

logical network on which information can be inferen-

ced. Commonly, the transactional systems in which

https://protege.stanford.edu/

http://www.hermit-reasoner.com/

the data acquisition of the academic ecosystem is car-

ried out in higher education institutions have two cha-

racteristics: the systems do not use a single reposi-

tory of information but are found in different sepa-

rate databases, and they use relational databases. The

management of multiple entity-relational models im-

plies that the data in each of these repositories must

be interpreted to derive their meaning, but the inter-

pretation lends itself to ambiguity. On the contrary, an

expressive language of knowledge such as ontologies

allow interpreting the knowledge unambiguously.

A strategy that can be used to transfer the infor-

mation stored in a traditional database to a semantic

network is through the data mapping procedure. Data

mapping allows populating the ontological network

with a semantic encoding of instances in RDF format

in an ontological scheme. As can be seen in Figure 3,

on the left side there is a fragment of the model entity

relationship of an academic system of students of a

higher education institution and on the right its equi-

valence in the model of the ontological ecosystem, in

which we can map the information between schemes

starting from their tables in the relational model:

• Student: The table contains the personal informa-

tion of the student, including ID, name and sur-

name, email and date of birth. The representa-

tion of a student in the ontological model separa-

tes concepts of the role as a student and the person

itself.

• Subject: The table contains the information of the

name of a subject offered. The mapping in the on-

tological model is similar to the table where there

is only one data property of the name of the sub-

ject.

• Tuition: It contains a ﬁeld with the offer date

and a one-to-many relationship from Student and

also from Subject. The ontological model for this

section already implies that in order to enroll a

student in a subject there must be a process that

participates as an intermediary –in this case, the

instance of the Course class must also be instanti-

ated.

This ﬁrst mapping exempliﬁed from the informa-

tion of a student named Peter Smith allows us to de-

ﬁne in a logical way a ﬁrst interpretation of the infor-

mation that is stored in an entity relationship model

and its equivalence in an ontological model as instan-

ces of the deﬁned classes in the modular schemes of

the ontological network of the academic ecosystem.

As there are several transactional systems in the

same academic ecosystem in higher education insti-

tutions, we can show an example of an excerpt from

the relationship model of the employee administration

KMIS 2018 - 10th International Conference on Knowledge Management and Information Sharing

164

Figure 3: Mapping scheme between the model entity rela-

tionship of the academic system against its version in an

ontological.

system in which a data mapping is applied in order to

transform it into instances to populate our ontologi-

cal network as presented in Figure 4, where we can

interpret tables of the relational database as follows:

• Employee: The table contains the personal infor-

mation of an employee, as for the Student in Fi-

gure 3. In the ontological model as well as in the

previous model, the appointment of professor is

taken as a position within an organization and the

person as something totally different, where per-

sonal information can be mapped in a similar way

as in Student.

• Institutional Position: It contains the informa-

tion of the positions that exist in the institution. It

is mapped in a similar way with a data property

with the name of the position, but we make expli-

cit that there is an organization to which that po-

sition is related, which in this case would be the

institution of higher education.

• Employee Institutional Position: It is an inter-

mediate table between Employee and Institutional

Position where the start and end date ﬁelds of the

position are also counted. In the ontological mo-

del, a time interval is assigned to the position node

by deﬁning the start date as the end date as instan-

ces.

The mapping of this second fragment of the rela-

tional database of the employee management system

demonstrates the integration of information from

another source of information under the same schema

of the ontological network, taking into account that

the employee for reasons of exempliﬁcation was cal-

led Peter Smith as in the mapping of the academic

system.

Figure 4: Mapping scheme between the model entity re-

lationship of the employee management system against its

version in an ontological model.

3.4 Axioms and Rules of Uniﬁcation of

Persons

The AEO-I module helps us to identify people who

have unique identiﬁers by means of axioms about the

classes and their relationships, taking advantage of the

fact that OWL allows us to enrich the meaning of the

properties of the objects ought certain axioms, one

of these axioms are the functional properties that are

those that, given an individual, by means of this pro-

perty the individual can be at most related to a single

individual (Horridge et al., 2004), for the case of our

module we have made functional the object property

”identiﬁer of” which helps us to discern if an indivi-

dual is the same as another, as we can see in Figure

5, an example is presented where, through the object

property ”identiﬁer of”, a node of the Identity Card

class is related to two people: Andrew J. M. and A.

Jone, which implies that Andrew JM and A. Jone are

the same individual, this logically implies that the two

nodes are merged through the ”sameAs” relationship

of OWL.

Figure 5: Example of the “identiﬁer of” functional property

of the AEO-I module.

The functional property would be a good option to

detect similarity of nodes in our ontological network

but this scenario is possible only if our instances of the

Semantic Representation of Information by Ontological Networks to Improve Knowledge Management in Higher Education

165

Figure 6: SPIN rule capture to ﬁnd individuals of Identity

Card class that are similar.

“Identity Card” class have already gone through a

process to ﬁnd their similarity. As we could see in the

mapping process with the example of Peter Smith this

person is represented in the system as two instances

p1 and p2 corresponding to the Person class, but in

turn both p1 and p2 have different IDs instantiated as

dni1 and dni2, so by not pointing to the same ID could

not operate the functional property.

To solve this problem, we must ﬁrst unify the in-

stances of the “Identity Card” class by means of ru-

les using the SPIN language as presented in Figure 6,

where we compare whether the data properties in the

instances of the “Identity Card” class.

When we establish the rules and deﬁne the axioms

we will use, we can discover information between the

instances of people who claim that one individual is

similar to another. Returning to our example where

we had two nodes p1 and p2 that represented Peter

Smith, as a ﬁrst step we analyzed their identiﬁcations

dni1 and dni2 by means of the rule deﬁned in SPIN

where they would be uniﬁed forming a logical level

as a single node and later through the Functional pro-

perty axiom of ”identiﬁer of” will infer that p1 and p2

are the same individual. From a general perspective

we have integrated the information of two different

transactional systems in the same semantic context,

but at the same time we also integrate the informa-

tion under the same context, in our case an actor can

have the information of his academic process, but also

coexist in the same ecosystem as an employee of the

university.

4 EXPERIMENTATION AND

RESULTS

After checking the correct functioning of our schemes

in the ontological network we have migrated the infor-

mation corresponding to two systems of a higher edu-

cation institution, the ﬁrst system contains informa-

tion corresponding to the students since 1993 and the

second corresponding to the personnel management,

each of these systems uses a traditional database of

Univerisdad Polit

ecnica Salesiana. The Karma tool

helps us to exploit what is exposed in the data map-

ping, as we can see in Figure 7. Based on the views of

the two transactional systems, we were able to trans-

form the information stored in the tables into an RDF

data structure that is consistent with our ontological

network.

Once the mapping process was ﬁnished we obtai-

ned two sets of instances in the form of RDF structu-

res:

• The ﬁrst set, coming from the migration of the

transaction management system of employees,

generated around 30,000 triples, which are instan-

ces of different classes of our ontological network

detailed in Table 2 where it can be seen that there

are about 2,000 employees in the system.

• The second set of data, corresponding to students,

has 99879 records of people as can be seen in the

breakdown of classes in Table 3. This set gene-

rated more than 2 million triples in our exported

ﬁle.

Table 2: Results of the Mapping of the Transactional Em-

ployee System.

Class Instances

foaf:Person 2236

vcard:Email 2089

aeo:Identity Card 2236

vcard:Name 2236

vcard:Individual 2236

Table 3: Results of the Mapping of the Transactional Stu-

dents System.

Class Instances

foaf:Person 99879

vcard:Email 78043

aeo:Identity Card 99879

vcard:Name 99879

vcard:Individual 99879

These RDF triplet ﬁles and the modules of our

ontological network were charged to the Allegro

Graph

triplet database, in order to be able to ma-

nage the amount of information that was generated

from the mapping and design processes of the ontolo-

gical network. To comply with the uniﬁcation of our

information through SPIN rules, we use a script of the

Top Braid Composer

tool called Motion SPARQL,

http://usc-isi-i2.github.io/karma/

https://franz.com/agraph/allegrograph/

https://www.topquadrant.com

KMIS 2018 - 10th International Conference on Knowledge Management and Information Sharing

166

Figure 7: Capture of data mapping using the Karma tool.

which allows us to fulﬁll the role of an ETL tool. Fi-

gure 8 shows the diagram of the script where the tri-

ples are ﬁrst imported through a connection to the tri-

plet databases; then, through a process of building tri-

ples from rules with SPIN, we generate the similarity

relationships through the OWL sameAs relationship

to unify the information of the IDs and subsequently

export the triples in a document in OWL format.

Figure 8: Capture of the SPARQL Motion script for gene-

ration of similar triples using SPIN.

Based on the rules with SPIN, we obtained that

out of a total of 102,115 people between employees

and students of the institution, 373 similarities of in-

dividuals belonging to the Identity Card class were

created, which means that there are this number of

people who are in the two systems, both that of stu-

dents and that of employees, but they were uniﬁed in

our ontological network, having the same context.

We loaded the set of pre-processed triples with

SPIN rules to the triplet database to populate our onto-

logical network, where by means of a SPARQL query

as shown in Figure 9, we were able to verify the uni-

ﬁcation of the people who had the same ID in our in-

stances of the triples:

The result of the SPARQL query is presented in

Table 4, where two columns are presented of which

each row is a similarity correspondence of instances

of the foaf class: Person, individuals with an IRI that

Figure 9: SPARQL query capture in AllegroGraph.

Table 4: Results of SPARQL consultation of people simila-

rity.

Node Person 1 Node Person 2

aeo:psd69 aeo:ps999

aeo:psd7 aeo:ps633

aeo:psd70 aeo:ps102219

aeo:psd71663 aeo:ps8889528

aeo:ps76 aeo:ps105536

starts with “psd” followed of a number are correspon-

ding to the instances generated from the employee ad-

ministration system and the instances that start with

”ps” followed by a number correspond to the student

system.

5 CONCLUSIONS

The present investigation addressed the construction

of an ontology that formalizes the knowledge of the

processes produced from the chair in institutions of

Semantic Representation of Information by Ontological Networks to Improve Knowledge Management in Higher Education

167

higher education, addressing its main actors, both te-

achers and students. Based on the NeOn methodo-

logy, we were able to cover the processes involved in

the life cycle of our formal model, based on an in-

terview and competency questions for the acquisition

of our requirements, which were the starting point for

the other processes of both resource reuse semantics

as well as the reengineering of non-ontological re-

sources, plus an alignment process to form our on-

tological network.

We have tackled some problems of the informa-

tion distributed in several databases of transactional

systems, where the interpretation of that information

lends itself to ambiguity, where by means of mapping

processes plus axioms and SPIN rules we can inte-

grate and unify more than twenty thousand records in

a set of instances that populated our ontological net-

work with 2 million triples that more than allows the

coexistence of information from different sources hel-

ped us to discern what 373 of employees and students

were the same people.

As future works it is proposed to be able to use this

ontological scheme in two directions: First, discover

relevant information that allows to manage in a more

efﬁcient way both the human and physical resources

of the University. Second to ﬁnd through the onto-

logical system characteristics that allow us to predict

the performance of a student. In this context it is also

very important to identify patterns related to students

that could lead to a state of student loss or desertion,

and be able to propose possible mitigating actions.

ACKNOWLEDGEMENTS

This work has been supported by the European Regi-

onal Development Fund (ERDF) and the Galician Re-

gional Government under agreement for funding the

AtlantTIC Research Center for Information and Com-

munication Technologies, as well as the Ministerio de

Educaci

on y Ciencia (Gobierno de Espa

na) research

project TIN2017-87604-R.

REFERENCES

Aminah, S., Afriyanti, I., and Krisnadhi, A. (2017).

Ontology-based approach for academic evaluation sy-

stem. In 2017 IEEE 33rd International Conference on

Data Engineering (ICDE), pages 1569–1574.

Arp, R., Smith, B., and Spear, A. D. (2015). Building Onto-

logies With Basic Formal Ontology. The MIT Press.

Gasmi, H. and Bouras, A. (2018). Ontology-Based Educa-

tion / Industry Collaboration System. IEEE Access,

6:1362–1371.

Horridge, M., Knublauch, H., Rector, A., Stevens, R., and

chester August 27, 2004. World, pages 0–117.

Iannella, R. and McKinney, J. (2014). vCard Ontology-

for describing People and Organizations. W3C Group

Note NOTE-vcard-rdf-20140522.

Kalemi, E. and Martiri, E. (2011). FOAF-Academic onto-

logy: A Vocabulary for the academic community. Pro-

ceedings - 3rd IEEE International Conference on In-

telligent Networking and Collaborative Systems, IN-

CoS 2011, pages 440–445.

Klyne, G. and Carroll, J. J. (2006). Resource description

framework (RDF): Concepts and abstract syntax.

Laouﬁ, A., Mouhim, S., Megder, E. H., and Cherkaoui, C.

(2011). An Ontology Based Architecture To Support

The Knowledge Management In Higher Education. In

Multimedia Computing and Systems (ICMCS), 2011

International Conference on, pages 1–6.

Melgar, A. and Quilca, A. (2016). An architecture for or-

ganizational memory systems in institutions of higher

education. 2016 11th Iberian Conference on Informa-

tion Systems and Technologies (CISTI), pages 1–6.

Mora-Arciniegas, M. B., Piedra, N., and Tenesaca-Luna,

G. A. (2017). Semantic Representation of Teaching

Planning , Pilot Experience at UTPL. In 2017 IEEE

37th Central America and Panama Convention (CON-

CAPAN XXXVII), pages 1–6.

arez-Figueroa, M., G

omez-P

erez, A., Motta, E., and

Gangemi, A. (2012a). Ontology Engineering in a Net-

worked World. Springer.

arez-Figueroa, M. C. and G

omez-P

erez, A. (2012).

Ontology Requirements Speciﬁcation, pages 93–106.

Springer Berlin Heidelberg, Berlin, Heidelberg.

arez-Figueroa, M. C., G

omez-P

erez, A., and Fern

andez-

opez, M. (2012b). The NeOn Methodology for On-

tology Engineering, pages 9–34. Springer Berlin Hei-

delberg, Berlin, Heidelberg.

KMIS 2018 - 10th International Conference on Knowledge Management and Information Sharing

168