Optimization of Knowledge Availability in an Institutional Repository
Filippo Eros Pani, Maria Ilaria Lunesu, Giulio Concas, Carlo Stara and Maria Pia Tilocca
Department of Electrics and Electronics Engineering, University of Cagliari, Piazza d'Armi, Cagliari, Italy
Keywords: Knowledge Management, Open Archive, Institutional Repository, Multimedia Content, Semantic Metadata.
Abstract: The Institutional Repositories (IRs) based on Open Archives represent one of the main free access tools for
the results of scientific research, and their diffusion is continuously growing. In fact, more and more
universities and research centers rely on IRs in order to optimize the management and diffusion of scientific
work. During an industrial project aimed to the creation of the Analytic Sound Archive of Sardinia, was
born the idea to create an Institutional Archive to solve the problems of organization and availability of
information. The Archive would contain a linguistically annotated corpus to preserve, enhance and support
the oral traditions in the Sardinian language. The distinctive feature of this work is the proposal of a new
approach for formalization and management of knowledge using the tool DSpace (a Knowledge
Management System typically used for the organization and management of text documents) to store,
manage and query the electronic corpus made of audio clips. In this specific case we worked with a group of
audio recordings in a corpus and linguistic information added to that corpus with annotations. The
customization of the structures and interfaces of the chosen KMS, to ensure the availability and sharing of
knowledge, are also closely tied to this research.
1 INTRODUCTION
Internet has become the most popular place for
integration, exchange and sharing of information.
The significant increase in its user base is flanked by
an even more significant increase in the amount and
types of available content, mainly due to the process
of digitization of existing information and the
creation of new information generated on the Net.
The problem of content availability and organization
depends on two strictly related issues: availability of
appropriate capabilities of indexing and retrieval
inside knowledge management tools, and the ability
of users to understand and use these features.
Experience suggests that the solution of this
problem can be found in the use of organized
schemas and relevant standardized metadata, or data
tiers, that allow us to describe, classify, and organize
basic information, allowing retrieval and use (Heery
and Patel, 2000) (Lagoze and Van de Sompel, 2003)
(Lunesu, Pani and Concas, 2011).
The trend to the standardization and
interoperability of the tools and system for archiving
highly structured information has brought, in recent
years, to establish institutional interoperable
repositories created on Open Archive Initiative
(OAI) architecture, based on standardized and
shared metadata schemas (Chopey, 2005)
(Solodovnik, 2011).
IRs have become the privileged place for the
dissemination and use of scientific knowledge, for
archiving and preservation of digital resources, as
well as for the exchange of metadata, in order to
promote and support the organization and
management of digital resources created by
institutions and their members. The Dublin Core
(DC) metadata schema represents a reliable support
to manage such contents, as it easily matches with
other metadata schemas, increasing the granularity
and refinement of their structures.
Building solid, well-structured and shared
metadata schemas as DC significantly improves the
classification and the availability of stored resources
while ensuring greater knowledge availability for
end users (Hutt and Riley, 2005). This is the
contexts where the “Analytic Sound Archive of
Sardinia” project belongs. The project aims to create
an IR that contains a linguistically annotated corpus
to preserve, enhance and support Sardinian language
oral traditions, especially improvised poetry.
The electronic corpus will be made of a
collection of audio clips, including poetry contests
called cantada, canto a chitarra, mutetu longu,
tenore, etc. The Sardinian samples are stored and
302
Eros Pani F., Ilaria Lunesu M., Concas G., Stara C. and Pia Tilocca M..
Optimization of Knowledge Availability in an Institutional Repository.
DOI: 10.5220/0004173203020306
In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2012), pages 302-306
ISBN: 978-989-8565-30-3
Copyright
c
2012 SCITEPRESS (Science and Technology Publications, Lda.)
recorded at different language levels, from
acoustics-phonetics up to linguistic and
paralinguistic levels. In compliance with OAI, this
corpus will be inserted into an open IR and therefore
made available to researchers of the Sardinian
language, as well as all individuals with an interest
for the matter. The proposed work therefore aims to
show a concrete solution for the realization of such
an archive based on the use of the open-source
software DSpace.
This document is organized as follows: Section 2
shows the development of institutional repositories
and the Dublin Core standard; in Section 3 we
present the approach we used to build the Analytic
Sound Archive of Sardinia; Section 4 shows the
description of the steps undertaken to customize the
tools we used. The last section hosts our final
observations about the work done and the possible
future.
2 INSTITUTIONAL
REPOSITORIES
Since the Nineties of the last century, a new
phenomenon has affected the process of scientific
communication and knowledge sharing: the
appearance and spread of digital repositories of
scientific contributions in order to make the
movement of information more "agile". In 1991,
Paul Ginsparg, at Los Alamos National Laboratory
(USA), paves the way to the arXiv, a repository of
works on Physics and Mathematics.
In June 1994 Stevan Harnad sent a “subversive”
proposal to the mailing list of the Virginia
Polytechnic Institute: they ought to share their ideas
through the contributions of self-archiving on the
internet, in order to communicate their results more
effectively.
A new kind of open archive begins to emerge:
the IR, supported and managed by an institution,
such as an university, which incorporates the
contributions of its researchers.
In October 1999 a group of researchers and
librarians in Santa Fe (USA) marked the turning
point: the rise of the OAI, essential to the
management of technical aspects such as protocols
and data exchange standards, localization and
subsequent retrieval of scientific contribution, and
software such as operating tools and for indexing.
The OAI consolidates the experience and the
previous techniques and, above all, embodies a sort
of “philosophical” awareness.
At the beginning of the new century, when the
archives have already opened and operational, the
expression “Open Access” is used for the first time
in a public document: Budapest Open Access
Initiative manifesto (2002). It suggested for the first
time to adopt both strategies, called
“complementary”, to encourage the spread of the
open access system: the "self-archiving", i.e.
archiving in institutional and disciplinary "open
electronic archives", of articles by researcher and
"open access journals," the new generation of
scientific open access journals.
3 OPTIMIZATION OF
KNOWLEDGE MANAGEMENT
SYSTEM
During an industrial project aimed to the creation of
the Analytic Sound Archive of Sardinia, the idea to
create an Institutional Archive to solve the problems
of organization and availability of information came
forth. There was, thus, the need to have an efficient
tool that could classify and store the vast amount of
knowledge contained in an electronic corpus of
Sardinian language, and that could, at the same time,
allow a high usability in terms of ease of reference
as well as ease of query and communication.
3.1 Formalization of Knowledge
In this context, knowledge is represented not only by
the texts of the corpus, but especially by the meta
language and linguistic annotations that enhance
them.
For each audio clip, a set of metainformation
describing the content is needed in order to enable
the search and retrieval of data by local author, title,
date of recording, to more particular features like
linguistic variety or singing type. Each audio clip is
also enhanced by a set of linguistic annotations.
The insertion of the audio clips in the chosen
KMS required the formalization of all the associated
metainformation in the form of a structured set of
metadata.
Linguists and musicologists working on the
Sardinian Linguistic Sound Archive chose a list of
possible annotation levels (syllable, tone,
morpheme, syntagm, accents, etc.), useful for both
linguistic and musical analysis of audio recordings.
OptimizationofKnowledgeAvailabilityinanInstitutionalRepository
303
3.2 Choice and Customization of the
KMS
DSpace, an open source software package developed
in 2000 in the context of a joint project of the
Massachussetts Institute of Technology with
Hewlett-Packard, provides all the necessary tools for
creation and management of an IR based on the
Open Access model. Such an IR can collect, store,
index, preserve and make accessible the information
output created by universities and research institutes
in a digital format.
DSpace is designed as a central storage facility
able to collect all kinds of content from the
community relating to the institution through a user
interface as simple and intuitive as possible. It can
collect various types of digital resources including
text, images, video, audio, articles and preprints,
technical reports, working papers, datasets, and
learning objects directly from the creators.
DSpace was chosen to realize the Analytic
Sound Archive of Sardinia as it fulfills all the
requirements asked by linguists and musicologists. It
is in fact completely customizable, supports natively
Qualified DC metadata schema and is compatible
with OAI with the support of OAI-PMH. The
proposed approach allows to insert the corpus and
the associated knowledge inside of DSpace, ensuring
the maintenance of its structure and the ability to
interrogate and update it easily by adding or
modifying its contents. Each text of the corpus is
inserted into a DSpace item so that it can be
uniquely associated with all of the metadata needed
for the linguistic analysis. The audio file contains the
registrations and the original files with the
annotations are loaded inside of the item as a
bitstream, while the metadata are stored in the
system database.
The first step consisted in the insertion of the
customization of new qualifiers for the Dublin Core
descriptive metadata representation and a new
scheme called "asas" for the representation of the
annotations. When inserting the corpus into DSpace
it was decided to create a specific item for each of
audio clip. It was therefore necessary to set the
release wizard offered by DSpace by changing the
specific XML file responsible for entry forms (input-
forms.xml). The descriptive metadata, identified by
researchers, such as title, author, type of song,
instrument, etc., and all metadata corresponding to
linguistic annotations (phono, morpheme, word,
etc..), was associated to each item, together with the
original file containing the audio recording and the
original file of annotations.
Figure 1: Customization of DSpace metadata's Register.
After the insertion of metadata, the interface was
customized by replacing the standard forms
provided by DSpace using modules specfically
designed to allow the creation of items and the
release of DC metadata according to the specific
needs of the project. The metadata on the
annotations were inserted instead using direct import
because the high number of occurrences for each
item made it difficult to enter them manually, as
shown by Hillman and Westbrooks (2004).
Finally, we proceeded to customize the search
interface of DSpace in order to adapt it to new
metadata and to the particular needs of the Analytic
Sound Archive of Sardinia. In essence, all metadata
corresponding to linguistic annotations needed to be
indexed in DSpace’s search engine so that we could
find a certain audio clip even through the search of
an associated record. Furthermore, some descriptive
metadata such as location, type of performer and
contribution were indexed to allow effective
searching that exploited the granularity of the
metadata.
3.2.1 Metadata Schemas
The metadata are stored and managed by DSpace
through a special tool, the Metadata Registry, where
the Qualifed Dublin Core schema is configured by
default. It can nevertheless be changed, and new
customized schemas can be added. The system
offers two ways to configure the register: one is the
graphic interface named Manakin, and the other can
be used by the terminal. Each of them has a specific
purpose. The first method allows an authorized user
to act on the diagrams through an easy and intuitive
web interface.
Once you create a schema, the metadata can be
added one at a time, with any qualifiers and related
notes. This feature is crucial for the updating and
maintenance of the system as it can make
adjustments quickly and easily without the
intervention of a computer expert. Likewise, you can
KEOD2012-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
304
choose the second solution where using a specific
command, the metadata schema expressed in XML
can be imported in the register according to a
specific syntax.
During the creation of the Sound Archive,
metadata was to be gradually defined and refined by
linguists and musicologists, so it was decided to
insert and manage it through the DSpace web
interface.
We obtained a customized Qualified DC schema
with new specific qualifiers and a new schema
“asas” with the metadata to use for the insertion of
linguistic annotations, as you can see in the Figure 2.
Figure 2: The new schema “asas”.
Once we reached the final version of the scheme,
however, it was decided to formalize the metadata in
XML so that should the system be reinstalled or
transferred, the register could be quickly configured
via the import metadata command.
3.2.2 Insertion of Metadata in the KMS
DSpace is preset to use, during the phase of the
insertion of an item, the Qualified Dublin Core
metadata schema but at the same time allows to
customize it, or to create new metadata schemas
according to the users and their requirements. The
first step in order to submit the metadata in DSpace
was to organize and structure them as metadata
schemas. The second step was, instead, to configure
the knowledge management system so that it could
be adapted to the particular chosen metadata
schema.
The Manakin graphic interface offers a tool to
manage the Metadata Register that allows to
customize the preconfigured Dublin Core schema in
a rapid and intuitive way.
In the Metadata Register can be entered all new
qualifiers with their related comments to obtain the
desired application profile, which coexist in the
Qualified Dublin Core standard with its qualifiers,
qualifiers and made for the specific project,
Figure 3: Interface personalization input metadata.
Analytical Sound Archive of Sardinia.
In DSpace this entity is called “item” and is
"built" with a wizard that, using the pre-configured
modules, allows to specify the values of the meta-
information to be formalized as metadata. These
modules are ready for the insertion of the metadata
belonging to the Qualified Dublin Core schema, but
in the case study they were fully customized to fit
the new specific qualifiers for the archive.
3.2.3 Research Modules
DSpace offers a powerful search interface that is
configured to use the main DC metadata (like title,
author, language) or the free full-text research as
parameter for the search.
The researchers asked for specific search criteria,
besides the standard ones, so you can search for
audio clips by place of recording, the participation to
a particular event, but also the annotations they
contain. The last criterion, in particular, is the most
important function for the study of the language
corpus as it allows to perform statistical analysis on
the text easily and quickly, without the need to use
specific and complicated software interrogation.
The search indexes were modified so that all
needed metadata were selected as criteria in the
search interface and information like place of
recording, performers’ information, the number
indicating the event place and all metadata
corresponding to the levels of annotation were
specifically added.
4 CONCLUSIONS
In the context of an industrial project that aimed to
create the Analytic Sound Archive of Sardinia, we
proposed a new approach using DSpace as a specific
tool to organize and manage a big quantity of
information coming from the audio recording, in
OptimizationofKnowledgeAvailabilityinanInstitutionalRepository
305
order to formalize knowledge in the Analytical
Sound Archive of Sardinia. We have chosen DSpace
because after some investigations we found that
more Universities and research Institutes (i. e.
Brunel University, Cornell University and
Massachusetts Institute of Technology) use DSpace,
in fact it is a very efficient tool easy to use,
customizable and flexible to allow the management,
the classification and the storage of a vast amount of
knowledge contained in an electronic corpus of
Sardinian language, and that could, at the same time,
allow a high usability in terms of ease of reference
as well as ease of query and communication.
REFERENCES
arXiv (Cornell University Library), http://arxiv.org/
Barton, J., Currier, S., Hey, J. M. N. (2003). Building
Quality Assurance into Metadata Creation: an
Analysis based on the Learning Objects and e-Prints
Communities of Practice. In Sutton, S., Greenberg, J.
and Tennis, J., Eds. Proceedings 2003 Dublin Core
Conference: Supporting Communities of Discourse
and Practice - Metadata Research and Applications,
Seattle, Washington (USA).
Becker, H., Chapman, A., Daviel, A., Kaye, K., Larsgaard,
M., Miller, Nebert, D., Prout, A. and Wolf, M. P.
(1997). Dublin Core element: Coverage. http://www.
alexandria.ucsb.edu/publicdocuments/metadata/dc_co
verage.html
Berlin Declaration on Open Access to Knowledge in the
Sciences and Humanities (2003). Retrieved from: http:
//oa.mpg.de/lang/en-uk/berlin-prozess/berliner-erklaru
ng/
Brunel University Research Archive (BURA), Brunel
University London, UK, http://dspace.brunel.ac.uk/
Chopey, M. A. (2005). Planning and Implementing a
Metadata-Driven Digital Repository. Haworth Press
Inc.
Cornell University Library (Cornell University, USA),
http://ecommons.library.cornell.edu/index.jsp
DSpace, http://www.dspace.org/
DSpace@MIT, MIT's institutional repository
(Massachusetts Institute of Technology, USA), http://d
space.mit.edu/
Dublin Core Metadata Initiative (DCMI).
Gartner, R. (2008). Metadata for Digital Libraries: state of
the art and future directions. JISC Technology and
Standards Watch Reports. Retrieved from: http://
www.jisc.ac.uk/media/documents/techwatch/tsw_0801
pdf.pdf JISC.
Heery, R. and Patel, M. (2000). Application profiles:
mixing and matching metadata schemas. Ariadne. http
://www.ariadne.ac.uk/issue25/app-profiles/
Hillmann, D. I. (2005). Using Dublin Core. Dublin Core
Metadata Initiative Recommendation. Retrieved from:
http://dublincore.org/documents/usageguide/
Hillman, D. I. and Westbrooks, E. L. (2004). Metadata in
practice. American Library Association.
Hutt, A. and Riley, J. (2005). Semantics and Syntax of
Dublin Core Usage in Open Archives Initiative Data.
Joint Conference on Digital Libraries.
Jackson, A. S., Han, M. J., Groetsch, K. and Mustafoff, M.
(2008). Dublin Core Metadata Harvested Through
OAI-PMH. In Proceedings of the 5th ACM/IEEE-CS
joint conference on Digital libraries.
Lagoze, C. and Van de Sompel, H. (2003). The making of
the Open Archives Initiative protocol for metadata
harvesting. Library Hi Tech.
Lunesu, M. I., Pani, F. E. and Concas, G. (2011). An
approach to manage semantic informations from UGC.
International Conference on Knowledge Engineering
and Ontology Development (KEOD).
Lunesu, M. I., Pani, F. E. and Concas, G. (2011). Using a
standards-based approach for a multimedia
knowledge-base. International Conference on
Knowledge Management and Information Sharing
(KMIS).
Open Archives Forum: OAI-PMH Tutorial, http://www.
oaforum.org/tutorial/
Open Archives Initiative, http://www.openarchives.org/
Solodovnik, I. (2011). Metadata issues in Digital
Libraries: key concepts and perspectives. Italian
Journal of Library and Information Science, Vol. 2,
No. 2.
KEOD2012-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
306
