Improving Knowledge Retrieval in Digital Libraries Applying
Intelligent Techniques
Antonio Martín and Carlos León
Polytechnic School, Sevilla University, Avda. Reina Mercedes, Seville, Spain
Keywords: e-learning Systems, Ontology, Case-based Reasoning, Knowledge Management, Intelligent Agents.
Abstract: Nowadays an enormous quantity of heterogeneous and distributed information is stored in the digital
University. Exploring online collections to find knowledge relevant to a user’s interests is a challenging
work. The artificial intelligence and Semantic Web provide a common framework that allows knowledge to
be shared and reused in an efficient way. In this work we propose a comprehensive approach for
discovering E-learning objects in large digital collections based on analysis of recorded semantic metadata
in those objects and the application of expert system technologies. We have used Case Based-Reasoning
methodology to develop a prototype for supporting efficient retrieval knowledge from online repositories.
We suggest a conceptual architecture for a semantic search engine. OntoUS is a collaborative effort that
proposes a new form of interaction between users and digital libraries, where the latter are adapted to users
and their surroundings.
1 INTRODUCTION
Digital repositories enable users to interact
effectively with information distributed across a
network: publications, forms, guides, educational
objects, legislation, etc. Access to these collections
poses a serious challenge, however, because present
search techniques based on manually annotated
metadata and linear replay of material selected by
the user do not scale effectively or efficiently to
large collections. In the traditional search engines
the information is treated as an ordinary database
that manages the contents. The result generated by
the current search engines is a list of Web addresses
that contain or treat the pattern. The useful
information buried under the useless information
cannot be discovered. It is disconcerting for the end
user and sometimes it takes a long time to search for
needed information. Despite large investments and
efforts have been made, there are still a lot of
unsolved problems.
Thus, it is necessary to develop new intelligent
and semantic models that offer more possibilities.
Ontologies assist the extraction of concepts from
unstructured textual documents and E-learning
objects by serving as a source of knowledge in large
digital libraries.
There is a lot of researches on applying these
new technologies into current information retrieval
systems, but no research addresses Artificial
Intelligence (AI) and semantic issues from the whole
life cycle and architecture point of view. The study
(Jimeno-Yepes, 2010) presents a system, which uses
an ontology query model to analyse the usefulness
of ontologies in effectively performing document
searches and proposes an algorithm to refine
ontologies for information retrieval tasks with
preliminary positive results. The study (Sasaki,
2005) presents a formulation and case studies of the
conditions for patenting content-based retrieval
processes in digital libraries, especially in image
libraries. (Diaz-Galiano, 2009) uses a medical
ontology to improve a Multimodal Information
Retrieval System by expanding the user's query with
medical terms. This study (Chen, 2008) combines
swarm intelligence and Web Services to transform a
conventional library system into an intelligent
library system with high integrity, usability,
correctness, and reliability software for readers. This
research (Toledo, 2011) proposes organizational
memory architecture, and annotation strategies
based on domain ontologies to retrieve information
through natural language queries.
Although search engines have developed
effective searches, information overload obstructs
precise searches. Our work differs from related
445
Martín A. and León C..
Improving Knowledge Retrieval in Digital Libraries Applying Intelligent Techniques.
DOI: 10.5220/0005159604450453
In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2014), pages 445-453
ISBN: 978-989-758-049-9
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
projects in that we build ontology-based contextual
profiles and we introduce an approach used
metadata-based in ontology search and expert
system technologies. This work presents an
intelligent approach for optimize a search engine in
the specific domain of university storehouses. It
incorporates Semantic Web and AI technologies to
enable not only precise location of public resources
but also the automatic or semi-automatic learning.
The contribution has been divided into next
sections. In the first section, short descriptions of
important aspects in digital library domain and
semantic interoperability, the research problems and
current work in it are reported. The second section
focuses on the Ontology design process and provides
a general overview about our prototype architecture.
Then we summarize its main components and
describe how can interact AI and Semantic Web to
enhancement a search engine. Finally we present the
results of our on going work on the adaptation of the
framework and we outline the future works.
2 DIGITAL LIBRARY ON E-
LEARNING DOMAIN
Repositories and digital archives are privileged area
for the application of innovative, knowledge
intensive services that provide a flexible and
efficient method for searching information and
guarantee the user with a set of results actually
related to his/her interest. These network
information systems support search and display of
items from organized collections. A Digital Library
(DL) is an electronic storage and access environment
for information and knowledge stored in the digital
format either locally in a library, in a group of
networked libraries, or at remote location. Reuse this
knowledge is an important area in this domain. The
Semantic Web provides a common framework that
allows knowledge to be shared and reused across
user’s community (Sure & Studer, 2005).
Seville University institutional repository is
dedicated to the production, maintenance, delivery,
and preservation of a wide range of high-quality
networked resources for scholars, and students at
University and elsewhere. DL means an integrated
set of services for capturing, cataloguing, storing,
searching, protecting, and retrieving knowledge. It
comprises digital collections, services, and
infrastructure to support lifelong learning, research,
scholarly communication, preservation, etc.
Our aim here is thus to contribute to a better
knowledge retrieval in the institutional repositories
field. This scheme is based on the next principles:
knowledge items are abstracted to a characterization
by metadada description which is used for further
processing. This begets new challenges to docent
community and motivates researchers to look for
intelligent information retrieval approach and
ontologies that search and/or filter information
automatically based on some higher level of
understanding are required. To reach these goals we
need to consider information interoperability. In
other words the capacity of different information
systems, applications and services to communicate,
share and interchange data, information and
knowledge in an effective and precise way, as well
as to integrate with other systems, applications and
services in order to deliver new electronic products
and services.
European initiatives, such as interoperability
between public services, require establishing
collaborative semantic repositories among public
and private sector organizations. In this paper we
study architecture of the search layer in this
particular dominium, a web-based catalogue for the
University of Seville. The hypothesis is that with a
case-based reasoning expert system and by
incorporating limited semantic knowledge, it is
possible to improve the effectiveness of an
information retrieval system (Sun and Finnie, 2004).
More specifically, the objective is to explore and
understand the requirements for rendering semantic
search in an institutional repository and investigate
from a search perspective possible intelligent
infrastructures form constructing decentralized
digital repositories where no global schema exists.
2.1 Interoperability Requirements
Particularly we require Semantic Interoperability,
which is one of the key elements of the programme
to support the set-up of the European E-Services In
June 2002; European heads of state adopted the
Europe Action Plan 2005 at the Seville summit. It
calls on the European Commission to issue an
agreed interoperability framework to support the
delivery of European E-Government services to
users and citizens. This recommends technical
policies and specifications for joining up public
administration information systems across the EU.
These aspects are the pillars to support the
European delivery of E-Services of the recently
adopted European Interoperability Framework (EIF)
(SEC, 2003) and its Spanish equivalent (MAP,
2014). European Institutions and Agencies should
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
446
use the European interoperability framework for
their operations with each other and with users,
enterprises and administrations in the respective EU
Member States (EIF, 2014).
The ISO/IEC 2382 Information Technology
Vocabulary defines the aspects of interoperability as
a general concept or approach cover technical,
semantic, and organisational issues, usually
referenced as interoperability layers, Figure 1.
Functional
Capability to reliably exchange information,
Sharing architecture and services
Syntactic
standard formatting for
machine-to-machine exchange of data
ORGANISATIONAL
Coordinated processes in which different organisations achieve a
previously agreed and mutually beneficial goal.
SEMANTIC
Precise meaning of exchanged information, which is
preserved and understood by all parties
TECNICAL
Planning of technical issues involved in linking computer
systems and services
Abstract
Concrete
Figure 1: Abstraction layers interoperability.
Interoperability is conceived on different main
abstraction levels:
Organisational interoperability level: processes,
defined as workflow sequences of tasks,
integrated in a service-oriented environment.
Technical interoperability level: signals, low-
level services and data transfer protocols.
Semantic interoperability level: information in
various shared knowledge representation
structures such as taxonomies, ontologies, or
topic maps. Semantic interoperability shared
vocabulary, and its associated links to an
ontology, which provides the basis for machine
interpretation and understanding of the logic of
the message.
Exchanging normalized data is a prerequisite for
semantic interoperability and refers to the packaging
and transmission mechanisms for data. Two or more
entities achieve interoperability when they are
capable of communicating and exchanging data,
which concerns to specified data formats and
communication protocols. In this study we have
focused our work in semantic interoperability
analysis. This area implies the collaboration of many
actors, such as local repositories. For this purpose
we use ontologies and semantic approach, which
enable reusing of existing domain knowledge and its
further retrieval, providing a contextual framework
enabling unambiguous communication of
information represented.
However, semantic interoperability problems
emerge as these organizations may differ in the
terms and meanings they use to communicate,
express their needs and describe resources they
make available to each other. We must bear in mind
that interoperability framework is, therefore, not a
static document and may have to be adapted over
time as technologies, standards and administrative
requirements change. In the next sections we
establish the base of all these aspects in our platform
OntoUS.
3 ONTOUS ARCHITECTURE
In order to support semantic retrieval knowledge in
university institutional repositories we develop a
prototype named OntoUS based on ontologies and
expert system technologies. OntoUS can be part of a
bigger framework of interacting global information
networks including e. g. other digital libraries,
scientific repositories and commercial providers, and
relies as much as possible on standards and existing
building blocks as well as be based on web
standards. The architecture of our system is shown
in Figure 2, which mainly includes three parts:
intelligent user interface, ontology knowledge base,
and the search engine.
Figure 2: System architecture of OntoUS.
The proposed architecture is based on our
approach to information retrieval in an efficient way
by means of metadata characterizations and domain
ontology inclusion. It implies to use ontology as
vocabulary to define complex, multi-relational case
structures to support the Case Based-Reasoning
(CBR) processes. Our system works comparing
objects that can be retrieved across heterogeneous
repositories and capturing a semantic view of the
world independent of data representation. OntoUS
system uses its internal knowledge bases and
inference mechanisms to process information about
the electronic resources in Seville University
repositories. Ontology knowledge base is the kernel
part for semantic retrieval information. At this stage
we consider to use ontology as vocabulary for
defining the case structure like attribute-value pairs.
ImprovingKnowledgeRetrievalinDigitalLibrariesApplyingIntelligentTechniques
447
Ontology is the knowledge structure, which identify
the concepts, property of concept, resources, and
relationships among them to enable share and reuse
of knowledge that are needed to acquire knowledge
in the specific search domain.
Metadata elements have been effectively used for
providing a richer representation of digital objects
and collections. The metadata descriptions of the
resources and repository objects (cases) are
abstracted from the details of their physical
representation and are stored in the Case Base.
Inference Engine contains a CBR component that
automatically searches for similar queries-answer
pairs based on the knowledge that the system
extracted from the questions text. Case Base has a
memory organization interface that assumes that
whole case-base can be read into memory for the
CBR to work with it. We used a CBR shell, software
that can be used to develop several applications that
require cased-based reasoning methodology. In this
work we analysed the CBR object-oriented
framework development environments JColibri
(GAIA, 2009). This framework work as open
software development environment and facilitate the
reuse of their design as well as implementations. The
CBR engine uses an evaluation function to calculate
the new case ranking, and the answered question
updates the query and the rankings in the displays.
The questions are ranked according to their potential
for retrieval and matching.
Advanced conversational user interface interacts
with users to solve a query, defined as the set of
questions selected and answered by the user during
conversation. Interface is designed and developed to
improve communication between humans and the
platform. In our system the user interacts with the
system to fill in the gaps to retrieve the right cases.
Also we have implemented a context interface,
which allows retrieving cases enough to satisfy a
SQL query.
3.1 Case-based Reasoning
CBR is widely discussed in the literature as a
technology for building information systems to
support knowledge management, where metadata
descriptions for characterizing knowledge items are
used. We have chosen the framework jColibri a
java-based configuration that supports the
development of knowledge intensive CBR
applications and help in the integration of ontology
in them. In our CBR application, problems are
described by metadata concerning desired
characteristics of an institutional resource, and the
solution to the problem is a pointer to a resource
described by metadata. These characterizations are
called cases and are stored in a Case Base. CBR case
data could be considered as a portion of the
knowledge (metadata) about an OntoUS object.
Every case contains both index with the association
terms of the ontology and the relation documents
residing on the repository network.
The development of a quite simple Case-Based
Reasoning application already involves a number of
steps, such as collecting case and background
knowledge, modelling a suitable case representation,
defining an accurate similarity measure,
implementing retrieval functionality, and
implementing user interfaces. Compared with other
AI approaches, CBR allows to reduce the effort
required for knowledge acquisition and
representation significantly, which is certainly one
of the major reasons for the commercial success of
CBR applications.
Case-Based Reasoning is a problem solving
paradigm that solves a new problem, in our case a
new search, by remembering a previous similar
situation and by reusing information and knowledge
of that situation. Reasoning cycle may be described
by the following steps processes, Figure 3:
C ase B ase
New Search
(New Case)
Retrieved
Case
Solved C ase
Tested
Repaired
Case
Learned
Case
Revise
R
e
u
s
e
R
e
t
a
i
n
Retrieve
Confirmed Solution
Suggested Solut i on
New Case
SIMILARITY
ADAPTATION
VERIFY
LEARNING
PROBL EM
Figure 3: Case Based Reasoning Cycle in OntoUS.
- Retrieval. The system retrieves the closest-
matching cases stored in a case base. Main focus of
methods in this category is to find similarity
between cases.
- Reuse: a complete design where case-based and
slot-based adaptation can be hooked is provided. If
appropriate, the validated solution is added to the
case for use in future problem solving.
- Revise the proposed solution if necessary. Since
the proposed result could be inadequate, this process
can correct the first proposed solution. It should be
noted that the differences in adaptation power
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
448
depend on how well the domain is understood.
- Retain the new solution as a part of a new case.
The solution is validated through feedback from the
user or the environment. This process enables CBR
to learn and create a new solution that should be
added to the knowledge base.
The Open Source JColibri1 system provides a
framework for building CBR systems based on state-
of-the-art Software Engineering techniques. Our
motivation for choosing jColibri framework is based
on a comparative analysis between it and other
frameworks, designed to facilitate the development
of CBR applications. jColibri enhances the other
CBR shells: CATCBR, CBR*Tools, IUCBRF,
Orenge, in several aspects: open source framework
and Java implementation, which is one of our main
requirements with respect to the easy integration in
the OntoUS system implemented in J2EE
environment.
Another decision criterion for our choice is
connected with the fact that jColibri affords the
opportunity to incorporate ontology in the CBR
application to use it for case representation and
content-based reasoning methods to assess the
similarity between them. By providing easy to use
model generation, data import, similarity modelling,
explanation, and testing functionality together with
comfortable graphical user interfaces, the tool
enables even CBR novices to rapidly create their
first CBR applications. Nevertheless, at the same
time it ensures enough flexibility to enable expert
users to implement advanced CBR applications.
3.2 Ontology Development
The main objective of our system is to improve the
modelling of a semantic coherence for allowing the
interoperability of different modules of
environments dedicated to digital university. We
have proposed to use ontology together with CBR in
the acquisition of an expert knowledge in the
specific domain. The primary information managed
in the OntoUS domain is metadata about
institutional resources, such as guides, publications,
forms, digital services, etc. We need a vocabulary of
concepts, resources and services for our information
system described in the scenario requires definitions
about the relationships between objects of discourse
and their attributes (Taniar and Wenny Rahayu,
2006). OntoUS project contains a collection of
codes, visualization tools, computing resources, and
data sets distributed across the grids, for which we
have developed a well-defined ontology using RDF
language. RDF is used to define the structure of the
metadata describing resources. Our ontology can be
regarded as triplet OntoSearch:={profile, collection,
source, relation) where profiles represent the user
kinds, collection contains all the services and
resources of the institutional repository and matter
cover the different information sources: electronic
services, official web pages, publications, guides,
etc., and relation is a set of relationships intended
primarily for standardization across ontologies. A
detailed picture of our effort in designing this
ontology is available in the Figure 4.
Figure 4: Ontological distributed environment.
This shows the high level classification of
classes to group together OntoUS resources as well
as things that are related with these resources.
Profile ontology includes several attributes like
Electronic_Resources, Digital_Collections,
Publication_Catalogue, Public_Services, etc. We
choose Protégé as our ontology editor, which
supports knowledge acquisition and knowledge base
development (Protégé, 2013). It is a powerful
development and knowledge-modelling tool with an
open architecture. Protégé uses OWL and RDF as
ontology language to establish semantic relations.
For the construction of the ontology of our system,
firstly we determine the domain and scope of the
ontology: Publications Catalogue, Web Sites,
Electronic Resources, etc. Also ontology must be
adapted to needs of user kinds. Second we
enumerate important terms in the ontology. It is
useful to write down a list of all terms we would like
either to make statements about or to explain to a
user. Then we define the classes and the class
hierarchy. The ontology and its sub-classes are
established according to the taxonomies profile.
In order to relate ontology classes to each other,
we defined our own meaningful properties for the
ontology and we defined a class hierarchy associated
with meaningful properties. Slots can have different
facets describing the value type, allowed values, the
number of the values (cardinality), and other
features of the values the slot can take. In the
ImprovingKnowledgeRetrievalinDigitalLibrariesApplyingIntelligentTechniques
449
following we give a short RDF description that
defined the concept of the user teacher that is a
subclass of Member_Community_University.
<rdf:Description rdf:about="#Teacher">
<rdfs:comment rdf:datatype=
"http://www.w3.org/2001/XMLSchema#string">
Teacher profile for affiliated colleges
</rdfs:comment>
<rdfs:subClassOf rdf:resource=
"#Members of the University community"/>
<rdf:type rdf:resource=
"http://www.w3.org/2002/07/owl#Class"/>
</rdf:Description>
The last step is to provide a conversational CBR
system to retrieve the requested metadata satisfying
a user query we need to add enough initial instances
and item instances to knowledge base. 10.000 cases
were collected for user profiles and their different
resources and services. This is sufficient for our
proof-of-concept demonstration, but would not be
sufficiently efficient to access large resource sets.
Each case contains a set of attributes concerning
both metadata and knowledge. However, our
prototype is currently being extended to enable
efficient retrieval directly from a database, which
will enable its use for large-scale sets of resources.
As a plus, domain specific rules defined by domain
experts (manually or by tools) can infer more
complex high-level semantic descriptions, for
example, by combining low-level features in local
repositories. On one hand, the rules can be used to
facilitate the task of resource annotation by deriving
additional metadata from existing ones.
We come to a process for addressing complex
relations between ontologies. As mentioned in
previous sections, relations among ontologies can be
composed as a form of declarative rules, which can
be further, handled in inference engines. In our
approach, we choose to use the Semantic Web Rule
Language (SWRL), which is based on a combination
of OWL DL and OWL Lite with the Case-Based
reasoning sublanguages, to compose declarative
search rules.
4 VISUAL INTERFACE TO
EARLIER RETRIEVAL
The understanding about digital libraries and
repositories is quite different according to its
specific users. OntoUS monitors user’s tasks,
anticipates search-based information needs, and
proactively provide users with relevant information.
The objective of profile intelligence has focused on
creating of user profiles: Staff, Administrator, and
Visitor. The user interface helps to user to build a
particular profile that contains his interest search
areas in the university repositories domain. In an
intelligence profile setting, people are surrounded by
intelligent interfaces merged, thus creating a
computing-capable environment with intelligent
communication and processing available to the user
by means of a simple, natural, and effortless human-
system interaction. The user enters query commands
and the system asks questions during the inference
process. Besides, the user will be able to solve new
searches for which he has not been instructed,
because the user profiles what he has learnt during
the previous searchers.
This configuration contains the user
requirements most typically described the relative
needs, tasks, and goals of the user for an individual
search. For this a statistical analysis has been done
to determine the importance values and establishing
specified user requirements. A schematic of the
architecture is show in Figure 5.
Figure 5: User Profiles, Graphical User interface.
The advantage of CBR is that users need only
input text partially describing the search and then the
system can assist in further complete the problem
description in an interactive conversation style. The
following guidelines for CBR design were proposed:
reuse questions, order context questions before detail
questions, eliminate questions that do not distinguish
cases, ask for only one thing in a question, and use a
similar, short number of questions per case. The user
begins the search devising the starting query. After
searching, some resources are returned as results.
The results include a list of web pages with titles, a
link to the page, and a short description showing
where the keywords have matched content within
the page.
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
450
5 RETRIEVAL PROCESS OF
SIMILAR CASES
CBR systems typically apply retrieval and matching
algorithms to a case base of past problem-solution
pairs. CBR is based on the intuition that new
searches are often similar to previously encountered
searches, and therefore, that past results may be
reused directly or through adaptation in the current
situation. Retrieval processes get back information
from the case library a set of potentially useful
cases, all of which partially match the new situation.
Retrieval process identifies the features of the case
with the most similar query. Our Inference Engine
contains the CBR component that automatically
searches for similar queries-answer pairs based on
the knowledge that the system extracted from the
questions text. The system uses similarity metrics to
find the best matching case. Similarity retrieval
expands the original query conditions, and generates
extended query conditions, which can be directly
used in knowledge, Figure 6.
Figure 6: Retrieval cases process.
The use of structured representations of cases
requires approaches for similarity assessment that
allow to compares two differently structured objects,
in particular, objects belonging to different object
classes. Retrieval strategy used in our system is
euclidean approach (Finnie and Zhaohao, 2002).
This approach involves the assessment of similarity
between stored cases and the new input case, based
on matching a weighted sum of features. Euclidean
distance is basis of many measures of similarity and
dissimilarity, which is usually the right measure for
comparing cases. The distance between ranking
vectors case
1
and case
2
is defined as follows:
similarity(Case
I
,Case
R
) (Case
i
Case
i
)
2
i1
n
(1)
Euclidean distance is the square root of the sum
of squared differences between corresponding
elements of the two vectors. In our system euclidean
distance is used to compare search results across
variables. Each row of the matrix is a vector of n
numbers, where n is the number of variables. We
evaluate the distance, or in this case the similarity
between any pair of rows. An important advantage
of similarity-cased retrieval is that if there is no case
that exactly matches the user’s requirements, this
can show the cases that are most similar to her
query.
6 EVALUATION AND PROOFS
Experiments have been carried out in order to
evaluate the effectiveness of run-time ontology
mapping. The main goal has been to check if the
mechanism of query formulation, assisted by an
agent, gives a suitable tool for augmenting the
number of significant documents, extracted from the
Seville University institutional repository, to be
stored in the CBR. For our experiments we
considered 150 users with different profiles. So that
we could establish a context for the users, they were
asked to at least start their essay before issuing any
queries to system. They were also asked to look
through all the results returned by OntoUS before
clicking on any result. In each experiment we report
the average rank of the user-clicked result for our
baseline system, Google and for our search engine
OntoUS. Then we calculated the rank for each
retrieval document by combining the various values
and comparing the total number of extracted
documents and documents consulted by the user,
Figure 7.
Figure 7: Comparative of valid pages percentage.
In our study domain we can observe the best
final ranking was obtained for our prototype OntoUS
and an interesting improvement over the
performance of Google. Our system performs
satisfactorily with about a 94.6% rate of success in
real cases. Another important aspect of the design
and implementation of an intelligent system is
determination of the degree of speed in the answer
that the system provides. During the
experimentation, heuristics and measures that are
commonly adopted in information retrieval have
been used, Figure 8.
While the users were performing these searches,
an application was continually running in the
background on the server, and capturing the content
of queries typed and the results of the searches.
ImprovingKnowledgeRetrievalinDigitalLibrariesApplyingIntelligentTechniques
451
Statistical analysis has been done to determine the
importance values in the results. Previous figure
shows a sample plot of these parameters that was
collected as a part of the experiment. We can
establish that OntoUS speed in our DL domain
improves the proceeding time and the average of the
traditional search engine.
Figure 8: OntoUS search analysis report.
The results for OntoUS are 21.5% better than
proceeding time and 14.9% better than executing
time searches/sec in the traditional search engines.
7 CONCLUSIONS
Our study addresses the main aspects of a Semantic
Web information retrieval system architecture trying
to answer the requirements of the next-generation
Semantic Web user. For this purpose we presented a
system based on ontology and AI architecture for
knowledge management in the Seville repositories.
This scheme is based on the next principle:
knowledge items are abstracted to a characterization
by metadata description witch is used for further
processing.
We have been working on the design of entirely
ontology-based structure of the case and the
development of our own reasoning methods in
jColibri to operate with it. It introduced a prototype
web-based CBR retrieval system OntoUS, which
operates on an RDF file store. Furthermore an
intelligent agent was illustrated for assisting the user
by suggesting improved ways to query the system on
the ground of the resources in Seville University
Repositories according to his own preferences,
which come to represent his interests. Finally the
study analyses the implementation results, and
evaluates the viability of our approaches in enabling
search in intelligent-based digital repositories.
Future work will concern the exploitation of
information coming from others institutional
repositories and digital services and further refine
the suggested queries, to extend the system to
provide another type of support, as well as to refine
and evaluate the system through user testing. It is
also necessary the development of an authoring tool
for user authentication, efficient ontology parsing
and real-life applications
REFERENCES
Chen, L., (2008). Design and implementation of intelligent
library system. Library Collections, Acquisitions, and
Technical Services, Vol. 32, Issues 3–4, 2008, 127-
141.
Diaz-Galiano, M. C., Martin-Valdivia, M. T. & Urena,
L.A. (2009), Query expansion with a medical ontology
to improve a multimodal information retrieval system.
Computers in Biology and Medicine, Vol. 39, Issue 4.
European Interoperability Framework (EIF) Version 2.
(2004) [cited 21-01-2014]; Available from:
ec.europa.eu/isa/strategy/doc/annex_ii_eif_en.pdf.
Finnie G and Sun, Z. (2002). Similarity and metrics in
case-based reasoning. Int J Intelligent Systems 17 (3),
273-287.
GAIA - Group for Artificial Intelligence Applications,
(2014). jCOLIBRI project, http://gaia.fdi.ucm.es/
grupo/projects/. Complutense University of Madrid.
Jimeno-Yepes, A., Berlanga-Llavori, R. & Rebholz-
Schuhmann, D. (2010). Ontology refinement for
improved information retrieval, Information
Processing & Managemen. Volume 46, Issue 4,
Semantic Annotations in Information Retrieval.
Ministerio de Administraciones Públicas (2014).
Aplicaciones utilizadas para el ejercicio de
potestades. Criterios de Seguridad, Normalización y
Conservación: 22/01/2014 Version. <http://
www.csi.map.es/csi/criterios/index.html>.
PROTÉGÉ, (2013). The Protégé Ontology Editor and
Knowledge Acquisition System. <http://
protege.stanford.edu/>.
Sasaki, H., & Kiyoki, Y. (2005). A formulation for
patenting content-based retrieval processes in digital
libraries, Information Processing &
Management, Volume 41, Issue 1, Pages 57-74.
SEC (2003) 801. Commission Staff Workking Paper:
linking up Europe, the importance of interoperability
for egovernment services europa.eu.int/ISPO/ida/
export/files/en/1523.pdf.
Sun, Z. and Finnie G. (2004): Intelligent Techniques in E-
Commerce: A Case-based Reasoning Perspective.
Heidelberg: Springer-Verlag.
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
452
Sure, Y., and Studer, R., (2005). Semantic Web
technologies for digital libraries. Library Management
Journal, Emerald, vol. 26.
Taniar, D., Wenny Rahayu, J. (2006). Web semantics and
ontology. Hershey, PA: Idea Group Pub, 2006.
Toledo, C. M. & Ale M. A. (2011). An Ontology-driven
Document Retrieval Strategy for Organizational
Knowledge Management Systems, Electronic Notes in
Theoretical Computer Science, Vol. 281, 21-34.
ImprovingKnowledgeRetrievalinDigitalLibrariesApplyingIntelligentTechniques
453
