The SemanticGov Project semantic-gov

Copyright 2005 Digital Enterprise Research Institute. All rights reserved.

www.deri.org

The SemanticGov Project www.semantic-gov.org

An EU Perspective: Providing Integrated Public Services to Citizens at the National and Pan-

European level with the use of Emerging Semantic Web Technologies

Vassillios [email protected]

Fifth Semantic Interoperability for E-Government Conference, October 10-11, 2006

2

Overview

I. Introduction

II. Modeling the government domain

III. Implementations - Pilots

3

I. Introduction

4

Purpose of the SemanticGov project

• To combine PA domain modelling with Semantic Web Service technologies (namely WSMO-WSML-WSMX) in order to create a Semantic Web Service enabled Public Service execution environment (namely WSMO-PA)

• To check the applicability of this execution environment in National and Pan-European Public Services

5

SemanticGov: Providing Integrated Public Services to Citizens at the National and Pan-European level with the use of Emerging Semantic Web Technologies.

SemanticGov aims at building the infrastructure (software, models, services, etc) necessary for enabling the offering of semantic web services by public administration (PA). Through this cutting edge infrastructure, SemanticGov will address longstanding challenges faced by public administrations such as achieving interoperability amongst PA agencies both within a country as well as amongst countries, easing the discovery of PA services by its customers, facilitating the execution of complex services often involving multiple PA agencies in interworkflows. More importantly, this infrastructure will exploit SemanticGov as an enabler for total reengineering of PA service provision and propose a paradigm shift of today’s modus operandi.

SemanticGov Project FP6-2004-IST-4-027517 is funded by the European Commission within the INFORMATION SOCIETY TECHNOLOGIES (IST) Programme.

Participants ListCentre for Research and Technology Hellas (CERTH) Coordinator GreeceNational University of Ireland – Digital Enterprise Research Institute (NUIG) Partner IrelandLeopold-Brazens University of Innsbruck (LFUI) Partner AustriaUniversity of Rome “La Sapienza”/DIS (UOR) Partner ItalyCAPGEMINI Partner The NetherlandsSOFTWARE AG Partner GermanyOntotext (ONTO) Partner BulgariaALTEC (ALTEC) Partner GreeceGreek Ministry of the Interior, Public Administration, and Decentralization (MOI) Partner GreeceRegion of Central Macedonia (RCM) Partner GreeceCity of Torino (Citta Di Torino) Partner Italy  

Starting date: January 1st, 2006Duration: 36 monthsTotal Budget: 4,375,000.00 €EU Funding: 2,720,000.00 € 

www.semantic-gov.org

6

Why Semantic Web Services in eGov?

Semantic Web Services eGovernment

Discovery•Need2Service Converter

Invocation PA specific distributed workflows:

• 1st Workflow: Gathering the service evidences

• 2nd Workflow: Service Execution - Collaboration with third parties

• 3rd Workflow: Consequence propagation

Composition

Execution

Monitoring • Enhancing Administrative Transparency

7

II. Modeling the government domain

8

Why Domain Models?

Domain Models

Domain Models

SOAsSOAs

Sem

anti

c W

ebS

eman

tic

Web

Semantic Web ServicesSemantic Web Services

MD

DM

DD

“…the third basic component of the Semantic Web, (is) collections of information called ontologies”

(T. B. Lee, J. Hadler, O.Lassila, The Semantic Web, Scientific America)

“SWSO is a core service ontology and is domain-independent. However, it is often necessary to represent domain-specific ontologies.”(SWSO specifications)

Model-driven development … represents a business-driven approach to software systems development that starts with a computation independent model (CIM) describing the business context and business requirements.”

(ATHENA NoE - Specification of a Basic Architecture Reference Model)

The more behavior you find in the services, the more likely you are to be robbing yourself of the benefits of a domain model. If all your logic is in services, you've robbed yourself blind.

(Martin Fowler)

9

Initiatives to model the PA domain

Initiative PA Relevance

Depth Domain Perspective View

1 The three spheres in eGovernance

High Low Support Operation (O), Policy (P), Service Provision (S)

Conceptual Holistic

2 Gartner Government Performance Framework

High Medium-Low

O, P, S Conceptual Holistic

3 Faceted Classification of PA High Medium-Low

O, S Conceptual Holistic

4 SAP Public Sector Solution Map Medium Medium-Low

O Conceptual Process

5 Government Process Classification Framework

High High O, P Conceptual Process

6 ONTOGOV Medium Medium S Technical Holistic

7 UK GCIM High Low S Conceptual Object

8 WebDG Ontologies Low Medium-High

S Technical Holistic

9 DIP eGovernment Ontology Low High O, S Conceptual Object

10 USA FEA Business Reference Model

High Medium-High

O, S Conceptual Process

11 USA FEA other models Low Medium-High

O, S Technical Holistic

10

II. Modeling the government domainGEA = Governance Enterprise Architecture

11

Governance Enterprise Architecture (GEA)

What is GEA• GEA is a generic top-level Enterprise Architecture for the

overall governance system

• From a MDD perspective, GEA is a Computation Independent Model (CIM) for the Public Administration domain

• GEA providesDomain Description Service Description

GEA provides a reusable top-level domain description, focusing on the HOW and WHAT columns of the Zachman Framework

GEA provides a reference PA service description

12

GEA – Zachman Framework

GEA Core

13

GEA foundations

• Enterprise Architecture

– Primarily Object (WHAT), Process (HOW) view

• Public Administration Theory

– For defining domain concepts

• Speech Act Theory

– Modeling the society-PA interaction as a discourse

– Identifying the Informative-Performative parts of service provision

– Types of Speech Acts - Types of PA services

– Types of Speech Acts - Types of IS interoperability

14

An example of SAT concepts in GEA

Determining classes of administrative acts

Declarations attempt to change the world

Assertives, statements that may be judged true or false because they aim to describe a state of affairs in the world.

Directives, attempts by the speaker to get the addressee to do something.

Commisives commit the speaker to some future course of action.

Expressives express a psychological state

Declarative

Assertive

Directive

- Optative Grant for a third child

- Imperative Smoking is prohibited

PA informs and certifies the existence and the truth of certain world states I certify you are Greek

PA changes the world I declare you married

PA directs society to certain states

Types of Speech ActsCore Functions

of Public Administration

15

GEA Overview

Governance System

Political SystemPolitical SystemAdministrative SystemAdministrative System

SocietySociety

Dialogue

Service Provision

Public Policy Formulation

16

Formulate Public Policy Area Provide Service Area

OBJECTS

AREA

PROCESSES

AREA

Formulate Public Policy Objects area

Formulate Public Policy Processes area

Provide Service Objects area

Provide Service Processes area

The GEA Framework

17

PUBLIC SERVICE OUTCOME

SERVICE PROVIDERSOCIETAL ENTITY

produces

receives provides

LAW

is governed by

OUTPUT CONSEQUENCEEFFECT

has type

PUBLIC ADMINISTRATION ENTITY

ROLE

has

SERVICE COLLABORATOR

EVIDENCE PROVIDER

CONCEQUENCE RECEIVER

participates in

PIECE OF EVIDENCE

PRECONDITION

EVIDENCE PLACEHOLDER

PURPOSE OF EVIDENCE

contains

validates

has

sets

uses

receives

INPUT

OTHER INPUT

PA Service Model

18

WSMO-WSML-WSMX

• Web Service Modelling…• … Ontology -> WSMO

- Conceptual model for SWS: goal, ontologies, mediators, services

• … Language -> WSML Ontology Language for SWS WSML Variants: WSML Core, WSML DL, WSML Rule, WSML Full

• … Execution Environment and Architecture -> WSMX Middle-ware platform for SWS

www.wsmo.org

19

WSMO Service Model

PA Service Model

WSMO-PA

Mapping GEA to WSMO

20

Mapping GEA to WSMO

• WSMO-PA Model Direct Mapping– WSMO Goal to GEA Goal– WSMO Service to GEA Service

• WSMO Precondition to GEA Precondition/Evidence• WSMO Postcondition to GEA Output/Consequence• WSMO Effect to GEA Effect• WSMO Assumption to GEA Evidence

• WSMO-PA Model Partial Mapping• WSMO Model

• GEA-PA Model– Actors– Needs– Consequence– Law

• WSMO Model– Shared Variables– Mediator– Ontology

21

WSML

Shared Variables

Mediators

WSMO-PA

22

Effect

Service Provider

Evidence

GEA Interface

WSMO-PA

23

II. Modeling the government domainPan-European Public Services

24

Motivation

The European Single Market

The European Citizenship

Pan-European Public Services (PEPS) are public services that may involve

• actors, and/or

• information and/or

• other services

from at least two Member States irrespective of the technology used.

IDABC Programme

Interoperable Delivery of European eGovernment Services to public Administrations, Business and Citizens

25

Pan-European Public Services

• a typology of Pan-European Public Services

• a typology of semantic conflicts in a cross-border environment

SemanticGov proposes:

26

A typology of Pan-European Public Services

• A client from Member State 1 executes a Service provided by a PA agency of Member State 2

• The Input/Output of a Service executed in Member State 1 comes from/will be sent to Member State 2

• PA agencies from different Member States participate in a common workflow

• The Consequences of a PEPS executed in Member State 1 should be communicated to Member State 2

27

Types of Semantic Interoperability Conflicts in Pan-European Public Service provision

Conflict Type Country A Description Country B

Data level

Data-value

Evidence APrecondition AEffect AConsequence A

Different meaningDifferent meaningIs not applicable or validIs not applicable or valid

Evidence A’Precondition A’Effect A’Consequence A’

Data representation Evidence value A Different format Evidence value A’

Data Units Evidence value A Expressed in different units Evidence value A’

Data precision Evidence value A Expressed in different values scale/grade Evidence value A’

Granularity

Object Property

Evidence A Conflicts of any type Evidence A’

ObjectEvidence Placeholder A

Conflicts of any type Evidence Placeholder A’

Schema Level

Naming

Service Provider AEvidence Placeholder A

Similar names/different services Or different names/similar servicesSimilar names/different usage Or different names/similar evidence

Service Provider A’Evidence Placeholder A’

Entity Identifier Client A Identified differently Client A’

Schema-isomorphismEvidence Placeholder A

Contain different set of evidence Evidence Placeholder A’

Generalization

Evidence Placeholder AClient AService Provider A

EP1 in one country = EP1+EP2 in anotherDifferent categorizations/groupsDifferent administrative and organizational structures

Evidence Placeholder A1’+A2’Client A’Service Provider A’

AggregationEvidence Placeholder A

Evidences aggregated differently Evidence Placeholder A’

Schematic discrepanciesEvidence Placeholder A

Similar evidences different names Evidence Placeholder A’

28

III. Implementations - Pilots

29

Cases Presentation

• Service Composition

• Need2Services

• Data Mediation

• SemanticGov - SWS Architecture

30

Service Composition

The evidence gathering workflow in PA services

31

Service CompositionMatching the Evidence Placeholder with the Service Output

32

Service CompositionWorkflow automatic composition

Enhanced UDDI registry implemented with the t-models mechanism (e.g. Paolucci).

33

Discovering Services

N2S Converter

NeedsServices

PA

PA

PA

PA

PA

PA

PA

Needs-aware

Service unaware

Service-aware

Needs unaware

34

Discovering Services

PREFIX gea: < http://localhost/GEA.owl #> PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>SELECT ?Service WHERE { ?Service rdf:type gea:GEA_Public_Service. ?Service gea:hasPADomain gea:Transportation. ?Service gea:hasEffectType gea: PromoteSocialWelfare.?Service gea:hasAdministrationLevel gea: Municipality_Level.?Service gea:hasClientType gea: Citizen.}

Services in the GEA ontology

SPARQL query for matching public services to properties

Matching Clients’ Profile-to-Services

The GEA ontology

35

Semantics Based Data MediationIdentifying equivalence amongst evidences for cross-country service provision

ID Card Number

Identification of a person

ID Card

SSN

Identification of a person

SSN certification

Country X

Country Z

36

Semantics Based Data Mediation

Identifying equivalence amongst evidences for cross-country service provision

<ruleml:imp><ruleml:_rlab ruleml:href="#TranslateCitizenIdentificationID"/><ruleml:_head>

<swrlx:datavaluedProperyAtom swrlx:property="http://research.altec.gr/OnarServer/Καταστατικο.owl#Αρ_Ταυτοτητας">

<ruleml:var>CitizenIdentificationID</ruleml:var></swrlx:datavaluedProperyAtom><swrlx:datavaluedProperyAtom

swrlx:property="http://research.altec.gr/OnarServer/Statuto.owl#Num_di_Passaporto"><ruleml:var>CitizenIdentificationID</ruleml:var>

</swrlx:datavaluedProperyAtom></ruleml:_head>

</ruleml:imp>

SWRL code for matching the Greek ID Card number to the Italian Passport Number

37

SemanticGov top-level Architecture

From Vitvar T., Kerrigan M., Overeem v. A., Peristeras V. and Tarabanis K.: Infrastructure for the Semantic Pan-European E-government Services, Proceedings of the 2006 AAAI Spring Symposium on The Semantic Web meets eGovernment (SWEG), Mar. 27-29, 2006, Stanford University, CA

38

Relevant Literature

• GEA related– Peristeras V., Tarabanis K., Reengineering the public administration modus operandi through the use of reference domain models and Semantic

Web Service technologies, Proceedings of the 2006 AAAI Spring Symposium on The Semantic Web meets eGovernment (SWEG), Mar. 27-29, 2006, Stanford University, California, USA

– Peristeras V., Tarabanis K., Providing pan-European e-government services with the use of semantic web services technologies: a generic process model in R. Traunmüller (Ed.): Electronic Government, 4th EGOV International Conference, DEXA 2005, Copenhagen, Denmark, 22 – 25 Αug. 2005, Proceedings: Lecture Notes in Computer Science vol. 3591 Springer 2005, pp. 226-236

– Peristeras V., Tarabanis K., Governance Enterprise Architecture (GEA): Domain Models for e-Governance, 6th ACM International Conference on Electronic Commerce (ICEC’04), Delft, The Netherlands, 25-27 October 2004  

• PEGS related– Peristeras V., Loutas N,. Tarabanis K., Semantic Interoperability in Pan-European Public Services, 40th HICSS, 4-10 Jan. 2007, Hawaii – Peristeras V., Tarabanis K., Knowledge Management Requirements for Pan-European Public Administration Service Delivery, in Maria A.

Wimmer (Ed.) Knowledge Management in Electronic Government, 4th IFIP International Working Conference, KMGov 2003, Rhodes, Greece, May 26-28, 2003, Proceedings. Lecture Notes in Computer Science 2645, Springer 2003, pp. 37-47

• WSMO related– www.wsmo.org

• WSMO-PA related– Xia Wang, Goudos S., Peristeras V., Vitvar T., Mocan A., Tarabanis K., WSMO-PA: Formal Specification of Public Administration Service Model

on Semantic Web Service Ontology, 40th HICSS, 4-10 Jan. 2007, Hawaii– Peristeras V., Goudos S., Vitvar T., Mocan A., Tarabanis K., Towards Semantic Web Services for Public Administration based on the Web

Service Modeling Ontology (WSMO) and the Governance Enterprise Architecture (GEA), 5th EGOV, DEXA 2006, Krakow, Poland.• Implementations-Pilots

– Goudos S., Peristeras V., Tarabanis K., Semantic Web Application for Public Administration using OWL for Public Domain Data Knowledge Representation, WSEAS Transactions on Information Science and Applications Journal (accepted for publication)

– Goudos S., Peristeras V., Tarabanis K., A Semantic Web Application for Matching a Citizen’s Profile to Entitled Public Services, ICEG 2006: The 2nd International Conference on e-Government, 12-13 October 2006, University of Pittsburgh, USA

– Goudos S., Peristeras V., Tarabanis K., Mapping Citizen Profiles to Public Administration Services Using Ontology Implementations of the Governance Enterprise Architecture (GEA) models, Semantic Web for eGovernment workshop, European Semantic Web Conference (ESWC) 2006, 11th-14th June, Budva, Montenegro

• General SemanticGov related– Vitvar T., Kerrigan M., Overeem v. A., Peristeras V. and Tarabanis K.: Infrastructure for the Semantic Pan-European E-government Services,

poster, Proceedings of the 2006 AAAI Spring Symposium on The Semantic Web meets eGovernment (SWEG), Mar. 27-29, 2006, Stanford University, CA, USA

– Vitvar T., Mocan A., Peristeras V. and Tarabanis K. (2006), Semantic-enabled Integration of Cross-border E-Government Services, poster at the 15th International World Wide Web Conference (WWW ’06), Workshop on eGovernment, 23 May 2006, Edinburg, Scotland

39

Current Status of Work

• The project is just finishing its initial phase (Conceptual Analysis – Requirements). It will run until Dec. 2008.

• We work on the top-level architecture to provide a full reference architecture for implementing SWS in PA environment

• We apply existing WSMO tools (e.g. WSMX, WSMO Studio, etc.). Our aim is to develop PA specific plug-ins (e.g. WSMO-PA Studio).

• We want to test the above, (a) in two real services as provided by a Greek and an Italian public agency, (b) in a set of laboratory cases

Copyright 2005 Digital Enterprise Research Institute. All rights reserved.

www.deri.org

The SemanticGov Project www.semantic-gov.org

An EU Perspective: Providing Integrated Public Services to Citizens at the National and Pan-

European level with the use of Emerging Semantic Web Technologies

Vassillios [email protected]

Fifth Semantic Interoperability for E-Government Conference, October 10-11, 2006