SEMANTIC LIFE EVENT ENVIRONMENT

FOR EGOVERNMENT SERVICES

Towards a Holistic Support to Host eGovernment Services in a Scalar Manner

Luis Álvarez Sabucedo and Luis Anido Rifón

Depto. of Telematics, University of Vigo, Spain

Keywords:

Interoperability, semantics, eGovernment, reference architecture.

Abstract:

The huge amount of resources that are being devoted to eGovernment projects makes possible the soaring of

new solutions. This onrush of contributions leads us to some shortcomings such as lack of interoperability or

poor scalability. To overcome this situation, the use of Life Events is proposed. Introduced in the paper, Life

Events are the cornerstone for a semantic proposal to provide eGovernmentservices in an organized manner.

A methodology for the use of Life Events and, also, the modelling of the entire system are presented.

1 INTRODUCTION

In the present moment, we are being witnesses of a

huge effort in eGovernment development. As ICTs

(Information and Communication Technologies) are

getting a broad use, Public Administrations (here af-

ter PAs) are getting into this tendency. But it is impor-

tant to bear in mind that eGovernment is not solely a

simple replacement of technology to provide a 24/7

service. Indeed, provision of eGovernment solutions

involves a huge effort in reengineering all processes

involved in public services. As a matter of fact, this

technology forces PAs to orient services toward a bet-

ter service by putting the citizen in the centre of all

provided operations.

A drawback of this trend is that interoperability is

reduced as a result of the data based, independent ap-

proach used to fulﬁl solutions. In order to overcome

this situation, this paper presents a platform where

eGovernment services can be provided using a se-

mantic approach based on Life Events. Life Events,

as discussed later on, are those events that play a role

a in the normal life of citizen and stimulate the needs

that compel citizens to interact with the administra-

tion.

The proposal presented in this paper takes care

of the provision of intelligent framework that using

a holistic approach can perform operations in an open

and interoperable platform. This one will be devel-

oped using a semantic support to provide an integral

approach where solutions from different PAs may be

integrated.

The rest of the paper is organized as follows.

Firstly, we will present the eGovernment state of the

art. Secondly, we introduce the problem we are deal-

ing with. Later on, major technological features we

are going to deal with are shown: semantics and Se-

mantic SOA, the architectural style used in the sys-

tem. The next section introduces the concept of Life

Event as it is going to be considered in the frame of

our proposal. Later on, the proposal for a software

design and the methodology to address the provision

of the system is provided. Finally, some conclusions

are yielded.

2 STATE OF THE ART

A large amount of national projects oriented toward

the provision of eGovernment services is available.

In the scope of the different countries, several ini-

tiatives must be underlined: SAGA (Standards und

Architekturen in eGovernment Anwendungen)(KBSt,

2005) in Germany, e-GIF (eGovernment Interoper-

ability Framework)(UK GovTalk, 2004) in United

Kingdom, ADEA (l’Agence pour le dèveloppe-

ment de l’administration électronique)(French Gov-

163

Álvarez Sabucedo L. and Anido Rifón L. (2007).

SEMANTIC LIFE EVENT ENVIRONMENT FOR EGOVERNMENT SERVICES - Towards a Holistic Support to Host eGovernment Services in a Scalar

Manner.

In Proceedings of the Third International Conference on Web Information Systems and Technologies - Society, e-Business and e-Gover nment /

e-Learning, pages 163-168

DOI: 10.5220/0001283701630168

 SciTePress

ernment, 2004) in France, EIF (European Interop-

erability Framework)(Enterprise and Unit, 2005) for

European solutions, FEAF (Federal Enterprise Ar-

chitecture Framework)(POPKIN Software, 2004) in

USA,. . .

There also a large amount of projects from non

national entities: OntoGov(OntoGov, 2007), Ter-

regov(TERREGOV, 2005), EPRI(Epr, 2005), EU-

Publi(Mariangela Contenti and Baldoni, 2004),QUA-

LEG(QUALEG, 2005), SmartGov(SmartGov, 2005),

eGovernment Good Practice Framework, FASME,

eGoia,. . .

3 THE PROBLEM

In order to overcome this drawback, a new formu-

lation for the problem is proposed. As, actually, all

interactions between the citizen are driven by the ex-

ercise of right or the fulﬁlling of a demand from the

administration, we propose the expression of services

provided of the administration in those terms. Thus,

it is possible to focus on what the citizen is requesting

and not on the PA itself as it happens many times.

To undertake this approach, instead of developing

services in a data layer from use cases expressed in

natural terms, the provision of a semantic description

of services is proposed in a scalar and reusable man-

ner. To model that idea, we deﬁne the so-called Life

Event(here after LE). This one is the concept used to

refer to any particular situation a citizen must deal

with and requires assistance, support or license from

a PA. We can considered as “life events” situations

such getting certiﬁcations, paying a ﬁne, getting mar-

ried, moving,...

As LEs have being modelled in the semantic layer

we are dealing the problem in higher layer of abstrac-

tion. This feature allows us to deal with the problem

in a more scalable way as we can use technologies

that provide us interesting features in that area. Be-

sides, it is possible to develop additional services to

provide added value services: advanced data mining,

automatic service composition,. . .

4 SEMANTIC WEB

Speaking about semantic data, we are addressing the

problem of dealing with not just data but information.

The aim for this discipline is the provision of infor-

mation understandable by machines. In the literature,

several deﬁnitions or approximations to the concept

of ontology are provided. A quite suitable deﬁnition

for ontology may be(Gruber, 1993): “ An ontology is

a formal, explicit speciﬁcation of a shared conceptu-

alization of a domain of interest. ”

By means of this deﬁnition we are addressing

an ontology as a way to put in a concrete way ab-

stract information about a certain domain by means

of machine-understandable data format.

We would also like to outline a key aspect of this

technology: the agreement. In fact, it does not matter

too much how good or rich is your deﬁnition of the

medium, but how general is the agreement you can get

around your proposal. To be able to work as intended,

we need a general agreement among our ontology.

To undertake the provision of an ontology differ-

ent languages(Gómez-Pérez et al., 2003) are possi-

ble. OWL (Ontology Web Language)(Web Ontology

Working Group, 2004) is the W3C Recommendation

that covers most of DAML+OIL and it is intended to

provide a fully functional way to express ontologies.

This technology is the chosen one for our proposal.

By using OWL, we are addressing a standard, solid

and interoperable platform for the provision of this

solution.

A different problem is the provision of a semantic

description of services. In this case, a solid frame-

work accepted for a general use is not available for

the community. We can outline several alternatives:

OWL-S (Ontology Web Language - Services)(OWL-

S Coalition, 2005), WSMO (Web Service Mod-

eling Object)(SDK WSMO working group, 2005),

WSDL-S (Web Service Description Language - Se-

mantic)(Akkiraju et al., 2007), . . .

5 SEMANTIC SERVICE

ORIENTED ARCHITECTURE

At ﬁrst glance, we can consider Service Oriented Ar-

chitecture, here after SOA, as an approach/method-

ology to the problem of providing an Architecture

Model. The gist of this proposal lays on the deﬁnition

and modelling of Services themselves. This philoso-

phy leads us to not pay attention to other details such

as the network or the format for the messages. From

this point of view, a service is a software resource ac-

cessible and deﬁned by means of some key concepts:

advertisement, service, contract, data model and pol-

icy. All of them together deﬁne and model the service

in a holistic manner.

This approach will provide interesting features

such as loosely coupled, platform independent and

protocol independent. Nowadays, the most common

support for system modelled under this style is Web

Services. However, any other middleware can be used

instead of the former.

WEBIST 2007 - International Conference on Web Information Systems and Technologies

164

Figure 1: SSOA Services.

Besides, we are prone to use Semantic Service

Oriented Architecture, here after, SSOA. So, SSOA,

in our proposal, is considered as an architectural style,

see (Bass et al., 2003), that uses advantages of seman-

tic applied to the concepts shown for SOA. In this

context, a semantic version of the usual elements of

the SOA architecture is used, see Figure 1.

6 OUR PROPOSAL

Taking into account the former considerations, a se-

mantic based deﬁnition for LE is established. These

elements are going to play a main role in our case and

they are expressed using semantic terms shared by the

whole system. The deﬁnition of a LE includes the fol-

lowing items:

Task Title for the considered operation. Folk-

sonomies can play an interesting role as they pro-

vide support for semi-automatic enhancements of

discovering services.

Description High level description of the desired op-

eration expressed in natural terms from the point

of view of the citizen.

Input documents As previously stated, all opera-

tions carried out by the administration require

some input document. As minimum, citizen is re-

quested to provide a signed form in order to in-

voke the operation. This element plays a role sim-

ilar to preconditions in some environments.

Scope We must identify the scope of the operation

(local, national, international,. . . ) where it must

to be recognized.

Output document Of course, as result of any per-

formed operation, the PA in charge must provide

an output expressed in terms of the ontology. This

information will be put together into one or sev-

eral documents. This output will vary its content

from the expected document (i.e., a certiﬁcation,

a license,. . . ) to information about the failure to

get the expected document.

Security Conditions This is intended to express the

conditions for the security mechanism involved

during the whole process. This includes the iden-

tiﬁcation of both parties, citizen and PAs, and also

the way is stored by any agent involved that could

be able to use it.

Cost This will express the amount you have to pay

for the requested operation and/or also the time it

will take for the completion of the operation.

These elements will be deﬁned using the power

of semantic expression that will allow us to provide

advanced services for discovering and orchestrating

them. Life Events can also be tagged using well-know

metadata standard already proposed and endorsed by

relevant organizations.

So, we propose the transformation of ﬁnal ser-

vices as they are requested into new LEs expressed

in terms of the semantic deﬁnition using the former

items presented.

6.1 Methodology

In order to transform natural or normal services to LE

expressed in the proposed terms, a simple methodol-

ogy is yielded:

1. Identify the particular problem you are dealing

with in terms of features and PAs involved.

2. Decompose the problem into several different

problems that may be resolved in a single step,

i.e., each step must produce as output a document

meaningful for the citizen.

3. For each procedure identiﬁed in previous step,

look for the input documents, scope and cost.

These ones must be expressed in terms of the on-

tology of the system. If required, the ontology

will be expanded.

4. Identify internal partial steps that the citizen could

be interested in. These steps usually involve inter-

nal documents that may carry no meaning for the

citizen but are relevant for the administration.

5. Identify all possible documents created as possi-

ble ﬁnal steps of the operation.

6. Update all services and agents that may be aware

of the new service.

SEMANTIC LIFE EVENT ENVIRONMENT FOR EGOVERNMENT SERVICES - Towards a Holistic Support to Host

eGovernment Services in a Scalar Manner

165

7 DEVELOPING THE SOFTWARE

The whole design of the system will be driven by LEs

as they were presented in former sections. Through

them, all components will be sketched and the system

built up. In the process of deﬁning a software archi-

tecture a solid methodology must be undertaken. In

our case, we will take advantage of: Uniﬁed Mod-

elling Process (UMP)(I. Jacobson et al., 1999) and

Bass contributions(Bass et al., 2003); Model Driven

Architecture (MDA) (OMG, 2007) and Semantic Ser-

vice Oriented Architecture (SSOA).

Eventually, a solid methodology is derived based

on the phases of MDA and using the interactive pro-

posed by the UMP to design a SSOA-based system.

Also, UML(OMG, 2005) is used to provide a formal

description of the outputs/inputs of each phase.

The process of designing the system is based on

the deﬁnition of a Reference Model that will drive the

entire proposal. This Reference Model, obtained from

the domain experience, will be transformed into a

Reference Architecture by mean of the UMP. The for-

mer is made of use cases that include the requirements

for the system. Applying an individual study to each

use case, a Reference Architecture, using the Bass ter-

minology, can be derived. This ﬁts with the idea of

PIM (Platform Independent Model) from the MDA

terminology and the Model of Analysis in the UMP.

Final implementations in a particular software sup-

port, also known as PSM (Platform Speciﬁc Model)

in MDA would turn out to be the functional solution

for real world according to the proposed methodol-

ogy. Similar schemes have also been tested success-

fully in other ﬁelds(Luis Anido et al., 2002).

7.1 Model of Reference

Model of Reference involves the deﬁnition of several

very high level features and properties that establish

the main lines of the project. One of the most impor-

tant features in that sense is the deﬁnition of agents.

In our case, the following ones are included:

Client Agent. This agent acts on the behalf of the

citizen that is actually requesting an operation. A

wide range of agents may be involved in the inter-

change of information. Agents will be responsible

for tasks related to composition and orchestration.

This element ﬁts in the proﬁle of the component

intelligent client in Figure 1.

Service providers. These elements are the ﬁnal re-

sponsible for the service provision. This role is

assigned to PAs. This component plays mainly

the role of the Service Provider in Figure 1.

Figure 4: Model of relations in the system.

Blue Page Server. In this architecture an element

to solve the problem of locating the server and

the proper service is mandatory. This component

plays the role of the Semantic Matcher in Figure

Security agent. A third part agent responsible for au-

diting and supervising operations that may require

additional security considerations.

Note that the formers perfectly ﬁt in the SSOA

philosophy. The review of the requirement from the

citizen and the administration leads us to the identiﬁ-

cation of up to 10 general uses cases as shown on Fig

2 and 3 that also are part of the Model of Reference.

The system will evolve and provide support to op-

erations according to the schema shown on Fig. 4. At

last, the ﬁnal aim of the citizen is to achieve the docu-

ment suitable in each case as documents are the legal

prove of a status or the fulﬁlling of an operation when

dealing with Public Administrations.

7.2 Deﬁning a PIM

In the context of our methodology, PIM is deﬁned

mainly in terms of an Analysis Model. The former

is the result of applying to each use case a process to

generate the UML design for Analysis Cases. Anal-

ysis Cases are deﬁned in terms of classiﬁers classes

(boundary, control and entities classes) and grouped

in packages that work cooperatively to achieve a ma-

jor goal, a use case.

This leads us to the identiﬁcation of several UML

diagrams containing a more in detail description of

the problem. One of them is included to illustrate

how this methodology works, see Fig. 5. Actually,

each Use Cases generates its own diagram. All of

them together sketch the ﬁnal result in an independent

way of the chosen software architecture. The entities

identiﬁed in this process (boundary, control and en-

tities classes) can be reused for several initial cases

and lead to a simpler implementation of the platform.

Actually, each entity can be assigned to an indepen-

WEBIST 2007 - International Conference on Web Information Systems and Technologies

166

Figure 2: Use case related to the citizen. Figure 3: Use Cases related to the blue page server.

dent group of experts for its implementation in further

steps.

In our case, up to 25 different classes were identi-

ﬁed to perform in a completed manner the use cases

identiﬁed in the former phase of our methodology.

7.3 Developing PSMs

To provide a real software solution that can success-

fully fulﬁll actual services, the actual software must

be developed. In order to achieve this ﬁnal goal,

we use as input the PIM from the previous step and

map them into new software components. These new

components are the so called PSM: Platform Speciﬁc

Model. The process to obtain them is the next step

according to the MDA methodology. Thus, in case of

need to deploy another software platform, the same

PIM can be used and applied to other constrains or

circumstances.

In the present case, to develop the system, a Java

based environment was used. To implement the se-

mantic features of the system, the chosen option was

Jena(Hewlett-Packard, 2005) due to the advantages

in managing for semantic queries. Services are im-

plemented under Web Services deﬁned by mean of

OWL-S ﬁles. One of the most complex features to

implement due to the current state of art in seman-

tics is the provision of the entity SearchEngine in

the use case SearchLifeEvent. In our case, to de-

velop this component we take advantage of ARQ for

Jena(Hewlett-Packard, 2006). This is an engine that

implements support for SPARQL(W3C, 2006), a pro-

tocol to execute semantic query against an ontology.

Finally, as result, an implementation for each identi-

ﬁed entity in the former step is provided.

8 CONCLUSION

This paper presents an approach to model and fulﬁll

services in the eGovernment domain. The main aim

is the presentation of a methodology based on the use

of semantics and SOA to drive the design and devel-

opment of a whole system. The result of the complete

process must a system that can provide support in an

open, scalar and adaptable manner to solutions in the

domain.

This approach ﬁts perfectly on a proposal based

on Life Events as main concept where the entire sys-

tem is based on. The use of Life Events, something

not new in the area, has been reviewed in this paper

and re-elaborated under a different conception to in-

clude semantic meaning and citizen centered view in

the context of this solution.

Also, as no proprietary solutions are used and only

open technologies are taken into consideration, this

environment allows the interchange of data and oper-

ation among different agents that may be developed

under the provided basis by third parties.

The ﬁnal status for the proposed architecture is

a common and global place where heterogeneous

agents may interact using the provided support.

ACKNOWLEDGEMENTS

We want to thank “Ministerio de Educación y Cien-

cia” for their partial support to this work under grant

(TIN2004-08367-C02-01).

SEMANTIC LIFE EVENT ENVIRONMENT FOR EGOVERNMENT SERVICES - Towards a Holistic Support to Host

eGovernment Services in a Scalar Manner

167

Figure 5: Analysis model for ExecuteLE.

REFERENCES

(2005). Epri. Web available. http://www.epri.org/.

Akkiraju, R. et al. (2007). Web service

semantics - wsdl-s. Web available.

http://www.w3.org/Submission/WSDL-S/.

Bass, L., Clements, P., and RickKazman (2003). Software

Architecture in Practice. Addison Wesley.

Enterprise, D. and Unit, I. I. (2005). Euro-

pean interoperability framework for pan-

european egovernment services. Web available.

http://europa.eu.int/idabc/en/document/2319/5644.

French Government (2004). Adea. Web available.

http://www.adae.gouv.fr/adele/.

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

(2003). Ontological Engineering. Springer.

Gruber, T. (1993). A translation approach to portable on-

tology speciﬁcations. Knowledge Acquisition, pages

199–220.

Hewlett-Packard (2005). Jena.

http://www.hpl.hp.com/semweb/.

Hewlett-Packard (2006). ARQ for Jena. Web available.

http://jena.sourceforge.net/ARQ/.

I. Jacobson, G. Book, and J. Rumbaugh (1999). The Uniﬁed

Software Development Process.

KBSt (2005). Saga. Web available.

http://www.kbst.bund.de/-,182/SAGA.htm.

Luis Anido, Juan Santos, Judith Rodríguez, and Manuel

Caeiro (2002). An MDA-based model for vertical

application task forces. An experience Report. Pro-

cededings of the Sixth International Enterprise Dis-

tributed Object Computing Conference’02.

Mariangela Contenti, Alessandro Termini, M. M. and Bal-

doni, R. (2004). Supporting inter-administratin co-

operation: the eu-publi.com approach. EGOV, pages

226–233. LNCS.

OMG (2005). Uniﬁed modeling language. Web available.

http://www.uml.org/.

OMG (2007). Omg model driven architectures. Web avail-

able. http://www.omg.org/mda/.

OntoGov (2007). Ontogov. Web available.

http://www.ontogov.com/.

OWL-S Coalition (2005). OWL-S: Semantic

Markup for Web Services. Web available.

http://www.daml.org/services/owl-s/1.1/.

POPKIN Software (2004). FEAF. Web available.

http://government.popkin.com/frameworks/feaf.htm.

QUALEG (2005). Qualeg. Web available.

http://www.qualeg.eupm.net/.

SDK WSMO working group (2005). Wsmo. Web available.

http://www.wsmo.org/TR/d2/v1.1/.

SmartGov (2005). Smartgov. Web available.

http://www.smartgov-project.org/.

TERREGOV (2005). Terregov. Web available.

http://www.terregov.eupm.net/my_spip/index.php.

UK GovTalk (2004). e-GIF. Web available.

http://www.govtalk.gov.uk/.

W3C (2006). SPARQL Query Language for RDF. Web

available. http://www.w3.org/TR/rdf-sparql-query/.

Web Ontology Working Group (2004). Web

ontology language. Web available.

http://www.w3.org/2004/OWL/.

WEBIST 2007 - International Conference on Web Information Systems and Technologies

168