SMART DIFFUSION OF GOVERNMENT CONTENTS
Elena Sánchez-Nielsen
1
, Francisco Chávez-Gutiérrez
2
1
Dpto. E.I.O. y Computación, La Laguna University,
38271 La Laguna, Spain
2
Parlamento de Canarias. C/Teobaldo Power, 7,
38002 S/C de Tenerife, Spain
Keywords: E-Government services, Smart podcasting, thesauri.
Abstract: Podcasting provides a new way for government agencies of any size to extend their reach and information
delivery to citizens. Currently, this technology is being used as an effective and inexpensive tool to deliver
audio and video content on mobile devices and personal computers. Their essence is focused on creating
multimedia content for audiences that want to listen/view when they want, where they want and how they
want. In this paper, we focus our vision on to distribute public hearings and sessions of parliaments in an
efficient and effective way to citizens. In order to provide customized content subscriptions according to
citizens’ preferences, we propose a semantic framework based on thesauri tools to describe the content of
audiovisual information. In particular, we introduce three new components in traditional podcasting
systems: (i) catalogue, (ii) fragmentation and (iii) customized delivery. Using these elements, we can
establish how to present podcasting results according to customized user content features and how to reduce
the size of multimedia segments to be downloaded and therefore, the quantity of information to be
processed. This paper describes how this framework is being built on plenary sessions of the Parliament of
the Canary Islands.
1 INTRODUCTION
Governments have made rapid progress worldwide
in embracing ICT technologies for e-government in
the past years. Examples of such improvements are:
e-government portals of information and procedure
services, points of access to EU legal documents, tax
services and virtual offices of allegations. At the
same time, the quantity and diversity of information
available from public government sources is now
quite large and increasing. Information available to
citizens includes institutional communication (the
official publication of laws and regulations),
information on opportunities and promotional
information. This significant amount of information
is currently managed by traditional access paradigms
that focus on retrieval of data using queries on
structured database systems and information
retrieval techniques. In order to enhance the access
to information, we focus our vision on how to make
this information more accessible to citizens. In
particular, our research efforts are focused on to
improve the efficiency and distribution of
parliamentary sessions and public hearings to
citizens, audience of business, nonprofits and
institutions by means of podcasting technology and
the Semantic Web paradigm.
Traditionally, Semantic Web paradigm has
become popular in fields such as information
integration, information retrieval on Internet and
knowledge management. In this paper, we propose
the use of Semantic Web paradigm and in more
detail, the use of thesauri tools to achieve podcasting
systems based on parliamentary contexts. In
particular, we propose the using of resources based
on thesauri, with the purpose of structuring, indexing
and providing personalized audiovisual content.
With this solution, we introduce a new approach to
inform and distribute information in an efficient and
effective way to citizens and other audiences that
consists of four processes: (i) encode and archive,
(ii) catalogue, (iii) fragmentation and (iv)
customized delivery. Using these components, we
can establish how to present podcasting results
according to specific user content features and at the
same time, we can establish how to reduce the size
of multimedia segments to be downloaded and
294
Sánchez-Nielsen E. and Chávez-Gutiérrez F. (2006).
SMART DIFFUSION OF GOVERNMENT CONTENTS.
In Proceedings of the International Conference on e-Business, pages 294-299
DOI: 10.5220/0001424902940299
Copyright
c
SciTePress
therefore, the quantity of information to be
processed by end-users. With this approach, citizens,
public authorities and other audiences can access
more quickly and at lover cost to the information
that they require. Moreover, this approach can be
extended to other domains of the Public
Administrations (PA’s), whose goal is focused on
the access to public information and reach of
citizens and other authorities.
The remainder of this paper is structured in the
following way. Section 2 gives an overview of
podcasting technology and Semantic Web paradigm.
Section 3 presents our content-awareness based
podcasting framework and Section 4 provides
concluding remarks and future work.
2 RELATED WORK
Podcasting is a recent and effective medium to
deliver syndicated Web content (audio/video data)
by content providers to consumer users. Interactions
between providers and consumers consist of a two-
step process. First, the content providers make
accessible audio or video files on an available
webserver, which are often referred to as one
episode of a podcast. Then, the content provider
acknowledges the existence of these files by
referencing them in another file known as the feed.
The feed is a machine-readable list of the episodes
which may be accessed. This list is usually
published in RSS format (RSS 2.0 Specification
n.d.) (Really Simple Syndication), which provides
other information, such as publish dates, titles, and
accompanying text descriptions of the series and
each of its episodes. The feed is typically limited to
a short list of the most recent episodes. Second, a
consumer user uses a software program called a
podcatcher with the purpose of determining the
location of the most recent episode and
automatically downloads it to the user's computer or
portable players. The downloaded episodes can then
be played and replayed at anytime.
Currently, the using of a subscription feed of
automatic delivery of new contents is what
distinguishes a podcast from a simple download or
real-time streaming. As a result, subscriptions to
podcasting allow users to collect programs from a
variety of sources for listening or viewing offline at
anytime and anywhere. However, no personalized
deliveries of specific contents have been included in
traditional podcasting systems. In this context,
Semantic Web (W3C: Semantic Web n.d.) paradigm
becomes a key feature for ensuring an appropriate
response to the user requests.
At present, the Semantic Web, the future of the
current Web, is an extended web of machine-
readable information and automated services that
amplify the Web far beyond current capabilities. The
explicit representation of the semantics underlying
data, pages and other Web resources will enable a
knowledge-based Web. The path to machine-
processable data is to make the data smarter.
There are four different stages from data with
minimal smarts to data embodied with enough
semantic information for machines to make
inferences about it:
Text and databases (pre-XML): this
initial stage corresponds to applications to
proprietary applications. Thus, the smarts
are in the application and not in the data.
XML documents for a single domain:
data, in this stage, achieves application
independence within a specific domain
using structured categories defined to
contain information about one aspect or
attribute (e.g. publisher, subjects) of an
information resource (e.g. book, document,
image). An example of this context would
be the personalized access to multi-version
XML document repositories in an
EGovernment scenario (Grandi et al, 2005,
p. 281-290).
Taxonomies and thesauri based
resources: in this stage, data are composed
from multiple domains and accurately
classified in a hierarchically structured
controlled vocabulary of terms that are used
to describe information resources.
Relationships and cross-references can be
used to relate and thus combine data. Thus,
data is now smart enough to easily
discovered and sensibly combined with
other data. Dynamic taxonomies (Sacco,
2000, p. 468-479) have been proposed as a
tool to solve information access and
dissemination needs of e-administrations
(Sacco, 2005, p. 261-268). Intelligent
thesauri such as WordNet (Fellbaum,
1998), an online lexical reference system,
can also be used to support automatic text
analysis and artificial intelligence
applications.
Ontologies and rules: in this stage, a
structural specification is used to express
complex relationships among concepts of a
SMART DIFFUSION OF GOVERNMENT CONTENTS
295
domain of interest. New data can be
inferred from existing data by logical rules.
Different languages can be used to describe
ontologies in a formal and explicit way
such as RDF (W3C: Rdf. n.d.), DAML
(The DARPA Agent Markup Language
Homepage: Daml. n.d.), OIL
(Ontoknowledge project: Oil n.d.), and
OWL (W3C: Web ontology language n.d.).
Figure 1 shows the four different stages of the
smart data continuum.
Figure 1: The smart data continuum.
3 SMART DIFFUSION OF
AUDIOVISUAL CONTENTS
Our aim is to enable a customized podcasting video
system to citizens in an e-government scenario
related to parliamentary interventions, which involve
different participants, debates, subjects to be
discussed, and initiatives to be approved. Plenary
sessions are the sessions to be included due to the
public character such is specified at article 71 of the
Parliament of Canary Islands Regulation
(Reglamento del Parlamento de Canarias n.d.). With
this aim, we present a framework that supports a
new model of video podcasting to only deliver the
fragments that end-users are interested. With this
approach, we provide two main features to citizens,
audience of business, nonprofits and institutions: (i)
minimizing the time of searching a specific subject
and (ii) reducing the size of data to be downloaded
and processed.
In order to enable smart distribution of
audiovisual contents of plenary sessions, we focus
our vision on three goals: (i) the use of podcasting
technology as mechanism of effective distribution of
plenary sessions to citizens and other audiences, (ii)
the use of Semantic Web paradigm with the purpose
of providing semantic meaning to parliamentary
videos and making available an appropriate response
to the user requests and (iii) the automation of
podcasting system by means of the use of a
thesaurus that comprises all the activity areas of
European Communities.
Figure 2 illustrates the resulting framework. This
framework is composed by the following modules:
3.1 Encode and Archive
Two essential aspects need to be addressed by the
encoding and archive module in order to provide
high quality video to end-users: (i) clarity pictures
and (ii) selection of adequate sound bitrates in order
to produce sound quality near to FM radio. This
situation involves high size of videos to be collected
in the storage system.
The contents of plenary sessions are digitalized
using Osprey video capture cards (Osprey video
n.d.) in a Windows Media System, which is
responsible of encoding and archiving the
audiovisual content to a digital format for a posterior
processing. Other systems equipped with audio and
video capture cards are also used to record the
different commissions’ sessions that can take place
at the same time.
Due to the increasing amount of multimedia
material which is recorded from plenary and
commissions’ sessions, an efficient storage,
management and administration of such material is
of growing importance. This goal is achieved by
using a storage area network (SAN). This storage
medium is a dedicated network that is separate from
LANs and WANs and it is used to interconnect the
storage-related resources that are connected to one
or more servers that contain the different
applications related to podcasting system, e-
government services and database systems.
The storage system consists of RAID storage
systems as storage peripherals, fibre channel
switches as interconnection equipment among all the
components of the storage network, servers, backup
devices to offload backup operations and interface
cards as components to connect servers to storage
network. Raid systems include data protection, fault
tolerance and high scalability.
3.2 Catalogue
Catalogue process is achieved by thesauri experts
who analyze the audiovisual content with the
purpose of adding semantic meaning and different
annotations. This process is carried out in real time
Ontologies and
Automated Reasoning
Taxonomies and
Thesauri based resources
XML Documents Using
Single Vocabularies
Text Documents and
Database Records
ICE-B 2006 - INTERNATIONAL CONFERENCE ON E-BUSINESS
296
while the sessions are taken place or more later
when the sessions have concluded.
The annotations made by human thesauri expert
identify the following aspects: (i) the name of the
person who takes part, (ii) parliamentary initiative
and (iii) the reason of the intervention. At the same
time, the thesaurus expert describes the conceptual
content that is taken place at every time by means of
a hierarchy of concepts that are described by
descriptors or terms. These descriptors are obtained
from a thesaurus that is a basic model of ontology
consisting of a database of hierarchical terms with
semantic and generic relationships.
All the annotations are stored at a database
system with other complementary information about
the plenary session included by the system such as
name of the session, date, and agenda and so on.
Multilingual Eurovoc thesaurus (The Office for
Official Publications of the European Communities
n.d.) is used by thesauri experts to catalogue in an
ambiguous fashion the different pieces of the
audiovisual content. Currently, Eurovoc thesaurus
comprises:
Descriptors: Those consist of words or
expressions which denote in unambiguous
fashion the constituent concepts of the
domain covered by the thesaurus (e.g.
implementation of the law).
Non-descriptors: consist of words or
expressions which in natural language
denote the same concept as a descriptor
(e.g. application of the law) or equivalent
concepts (e.g. enforcement of the law,
validity of the law).
Semantic relationships: relationships based
on meaning, firstly between descriptors and
non-descriptors and secondly between
descriptors.
In more detail, Eurovoc thesaurus comprises 21
fields, 127 microthesauri, 6439 descriptors, 6448
hierarchical relationship and 3501 associate
relationship. The European Parliament, the Office
for Official Publications of the European
Communities, the national and regional parliaments
in Europe, some national government departments
and European organisations are currently using this
thesaurus.
The 21 different fields covered by this thesaurus
comprises all the areas of importance for the
activities of the European institutions: politics,
international relations, European Communities, law,
economics, trade, finance, social questions,
education and communications, science, business
and competition, employment and working
conditions, transport, environment, agriculture,
forestry and fisheries, agri-foodstuffs, production,
technology and research, energy, industry,
geography and international organizations.
Every different subject of the audiovisual
content is identified by a start and end tag. The
thesaurus expert describes in unambiguous fashion
the conceptual content between both tags using the
descriptors from Eurovoc thesaurus that most clearly
matches the conceptual concept. Each descriptor
uses a two-tier hierarchical classification that
comprises (i) a field, identified by two-digit numbers
and titles in words, e.g.: 10 European Communities
and (ii) a microthesauri, identified by four digit
numbers, e.g.: 1011 Community Law, where the two
first digits correspond to the field that contains the
microthesaurus. Semantic relationships between
descriptors comprise scope notes (some descriptors
can be accompanied by notes, clarifying the
meaning of the descriptor), microthesaurus
relationships, equivalence relationships, hierarchical
Encoding and
Archive
Thesaurus
expert
Catalogue
Video
EUROVOC
Deliver
User
Web page
Fragmentation
Video
channels
Figure 2: Smart diffusion of audiovisual contents
Figure 2: Smart diffusion of audiovisual contents.
SMART DIFFUSION OF GOVERNMENT CONTENTS
297
relationships and associate relationships.
Figure 3 illustrates graphically some fields of
activity of Eurovoc thesaurus and in more detail the
different descriptors associated to concept related to
“organization of work and working conditions”.
3.3 Fragmentation
After the catalogue process has been performed, the
annotations made by thesauri experts allow
identifying the start and end of the different subjects
discussed.
Each video of each plenary session is divided by
an automatic process in smaller videos that will have
the size equivalent to the subject discussed.
Once the fragmentation process has been
finished, each video fragment is assigned to the
corresponding channel. Each fragment of video
corresponds to one or several fields of activity of
Eurovoc thesaurus. These fields of activity are
computed by means of recovering the descriptors
assigned by thesauri experts at catalogue process and
by using hierarchical structure of the thesaurus
Eurovoc with the purpose of identifying the root
layer where this descriptor is associated.
3.4 Customized Delivery
The 21 different fields of Eurovoc thesaurus define
the available channels that can be syndicated by
citizens and other audiences.
In order to citizens know the available channels;
a list with the different channels will be published in
a web page. Each channel represents a feed, that is, a
XML file based on RSS specification (RSS 2.0
Specification n.d.). In particular, this specification is
the format used in podcasting that allows
transforming the way millions of people consumes
news and information. The growth adoption rate of
RSS allows people rarely read information directly
from a website.
In our podcasting architecture, each day, the
system will be responsible of generating the RSS file
of each channel in order to reflect the availability of
the new episodes. The 10 last episodes of each
channel will be included. In this context, users may
select the appropriated RSS files corresponding to
the channels they are interested.
Once syndication to a channel has been
achieved, the software program (podcatcher) of
every citizen or institution subscribed computes the
localization of the most recent episodes of video for
each user subscribed and automatically will
download it to the user’s computer or portable
player. The downloaded episodes can then be
(a)
(b)
Figure 3: Eurovoc thesaurus: a) Some main fields of activity of European institutions and b) Microthesaurus
associated to the field “Employment and Working Conditions
Figure 3: Eurovoc theasurus_ a) Some main fields of activity of European institutions and b) Microthesaurus associated to
the field “Employment and Working Conditions”.
ICE-B 2006 - INTERNATIONAL CONFERENCE ON E-BUSINESS
298
played, replayed, or archived as any other computer
file.
4 CONCLUSIONS AND FUTURE
WORK
In this paper, we present a new electronic service
delivery to citizens, audience of business, nonprofits
and institutions. We focus our vision on how to
make information based on audiovisual contents
more accessible to several audiences. In particular,
our research efforts are focused on to improve the
efficiency and distribution of public hearings and
parliamentary sessions by means of podcasting
technology and the Semantic Web paradigm.
Podcasting technology is used as a mechanism to
distribute audiovisual content, which can be played
and replayed offline at anytime. Semantic Web
paradigm gives a promising solution to customized
delivery. The explicit representation of the semantics
underlying podcasting system enables a qualitatively
new level of service and provides a new approach to
create smart podcasting to deliver customized
information to different audiences and automate the
assignation of information to different channels in
the podcasting system.
Currently, we have developed encode and archive
and catalogue process.
Future work will be focused on developing the
remainder processes and extending the framework
proposed in order to incorporate query formulation
by end-users who want to search the spoken words
related to main topics within any audio or video file.
Also, new optimization techniques that produce
video content according to different audience
players will be evaluated.
REFERENCES
Grandi, F, Mandreoli F, Martoglia R, Ronchetti E, Scalas
MR and Tiberio P, 2005. Personalized access to multi-
version Norm Texts in an eGovernment scenario,
M.A. Wimmer et al. (EDs.): EGOV 2005, LNCS
3591.
Sacco, G.M, 2000. Dynamic Taxonomies: A Model for
Large Information Bases. IEEE Transactions on
Knowledge and Data Engineering 12, 2.
Sacco G.M, 2005. Guided Interactive Information Access
for E-Citizens. M.A. Wimmer et al. (EDs.): EGOV
2005, LNCS 3591.
Fellbaum, C. (ed.), 1998. WordNet - An Electronic Lexical
Database. MIT Press.
W3C: Rdf. Retrieved from http://www.w3.org/RDF/
The DARPA Agent Markup Language Homepage: Daml.
Retrieved from http://www.daml.org/
Ontoknowledge project: Oil. Retrieved from
http://www.ontoknowledge.org/oil/
W3C: Web ontology language. Retrieved from
http://www.w3.org/2004/OWL/
RSS 2.0 Specification. Retrieved from
http://blogs.law.harvard.edu/tech/rss
The Office for Official Publications of the European
Communities. Eurovoc Thesaurus. Retrieved from
http://europa.eu.int/celex/eurovoc/
W3C: Semantic Web. Retrieved from
http://www.w3.org/2001/sw/
Reglamento del Parlamento de Canarias. Retrieved from
http://www.parcan.es/pub/reglamento.pdf
Osprey video. Retrieved from
http://www.viewcast.com/products/osprey.html
SMART DIFFUSION OF GOVERNMENT CONTENTS
299