Integrating Distributed Data Bases in a Semantic Framework

The K-Metropolis Project

Daniela Giordano, Alfredo Torre, Salvatore Alessi, Alfio Costanzo and Alberto Faro

Department of Electrical, Electronics and Computer Engineering,

University of Catania, viale A. Doria 6, 95125, Catania, Italy

Keywords: e-Government, Semantic Web, Location based Services, Distributed Database Systems.

Abstract: This paper presents how data integration is obtained in K-Metropolis, i.e. a project inspired by the

Connected Government framework and supported by the Regional Government of Sicily that aims at

integrating data residing on personal, commercial and municipal databases of different organizations. In

response to the user queries, K-Metropolis suggests the most suitable services by a decision support system

illustrated in previous works using all the data available at urban level. E-payment of the recommended

commercial services is supported. On request, e-government certificates are provided to the mobile users.

When needed, the data and certificates are displayed on a Google Maps based interface. In a companion

paper we show how K-Metropolis is able to display on a more powerful GIS not only geo-referenced data of

the public services but also thematic drawings that further qualify these data.

1 INTRODUCTION

In the last decade, the adoption of semantic

technologies for facilitating data interoperability has

been a major endeavour for the most of the

companies employing information systems

interrelated to data of other companies, typically

with the goal to support enterprise market needs or

to track goods for food origin control, e.g. (Sheth,

2005) and (Salampasis et al., 2005).

Recently, also Public Administrations (PAs) are

more and more interested in integrating all their

archives to better plan and control the economic and

social activities developing at either regional level or

urban scale, e.g., (Zhai et al., 2008) and (Costanzo et

al., 2012). In all these systems, individual citizens

and resources are represented by Uniform Resources

Identifiers (URIs) that are linked together by

properties, thus giving rise to a subject-predicate-

object graph, aiming at describing individuals or

objects by their distinctive features, whereas

knowledge needed for the data integration process is

encoded in ontologies.

In addition to data integration, achievable

through ontologies mappings to manage the data

semantic heterogeneity, a significant advantage in

resorting to semantic web technologies is the

inherently flexible data model upon which the

information system rests. The model does not need

to be complete and fixed, rather it remains open to

changes and refinements, also related to unforeseen

ways of consuming data (e.g. the type of queries to

be issued).

In fact, novel properties can be easily integrated

in the subject-predicate-object graph, that is by its

very nature open-ended. This aspect is especially

important in highly dynamic scenarios, with

changing regulations and policies and, in general,

where both structured, semi-structured and

unstructured data should coexist.

Traditionally, the challenge of data integration

has been tackled with Data WareHousing (DWH)

technologies; currently, there is a trend towards

leveraging on semantic web technologies for more

agile, incremental integration of data, thus

overcoming the inflexibility of the traditional DWH

model and the boundaries of the single organization.

First examples of DWHs of semantic data are

now being discussed, e.g., (Nebot and Berlanga,

2012), to retain the advantage that a DWH affords

for performing On-Line Analytical Processing

(OLAP) analysis of semantic data and combining it

with the inference power of the annotation

semantics. On the other hand, data and services

integration issues affect also Geographic

Information Systems (GIS) technology, whose key

role in decision making in the public sector is well

322

Giordano D., Torre A., Alessi S., Costanzo A. and Faro A..

Integrating Distributed Data Bases in a Semantic Framework - The K-Metropolis Project.

DOI: 10.5220/0004558203220328

In Proceedings of the 15th International Conference on Enterprise Information Systems (ICEIS-2013), pages 322-328

ISBN: 978-989-8565-59-4

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

established. It has been recently remarked that

although much research has been done on the

integration between DWH, OLAP and GIS, the main

shortcoming of the proposed solutions is that they

are not open and extensible, whereas the design of

the spatial dimensional schemas and of meta-models

for the semantic integration among the metadata of

such technologies is still problematic, e.g., (Nebot

and Berlanga, 2012).

Effective solutions for the integration of all of

the above technologies from a semantic web

standpoint are therefore topical, since they will be at

the core of the new breed of unified information

systems needed by advanced e-government and

“smart cities” applications that are evolving having

in mind the Connected Government framework

whose aim is “to shift from a model of providing

government services via traditional modes to

integrated electronic modes wherein the value to the

citizens and businesses gets enhanced” (Saha, 2010).

This work presents how the mentioned data

integration is obtained in K-Metropolis, i.e. a project

supported by the Regional Government of Sicily, to

collect data originating from personal, commercial

and municipal databases of different organizations to

offer suitable e-services to citizens and businesses.

When needed, such data are displayed on a Google

Maps based interface, whereas the use of a more

powerful GIS is discussed in a companion paper

(Giordano et al., 2013). Sect.2 illustrates the

functional and implementation architecture of K-

Metropolis, whereas sect.3 sketches the solution

adopted for data integration and the user interface to

provide the needed location based information

services to desktop PCs and user mobiles.

2 THE K-METROPOLIS

PROJECT

The K-Metropolis project focuses on data

integration to support urban/metropolitan activities

from two main points of view: i) logistics activities

and personal mobility based on traffic ontologies,

e.g. (Faro et al., 2003), and ii) management of the

municipal data bases to collect taxes and to improve

the services offered to the citizens. Concerning the

first point, K-Metropolis plans to use sensing

infrastructures, mainly based on computer vision

techniques, e.g. (Faro et al., 2008, 2011a, 2011b)

and (Crisafi et al., 2008), to support mobility and

logistics information in real-time. Real-time

provision is also a requirement taken into account to

design data integration to obtain the certificates on

citizens’ demands sent from fixed and mobile

devices.

The available proposals for distributed e-

government applications typically consist in the

integration of several data bases coded in proprietary

format. On the contrary, in the K-Metropolis project,

the data are organized in a sort of “Location-aware”

Data Warehouse (LA-DWH) where semantic

relations link citizens activities to services, e.g.,

finding and reserving “nearest to me” services such

as parks, pharmacies and fuel stations, or on-line

issuing of personal certifications.

By this choice, the main ingredients of the above

sketched urban/metropolitan information system,

i.e., the activities and service flows involving

citizens and the relevant location (street, apartment,

office, stores, parks, etc.), can be identified by stable

URIs such as codes and addresses (e.g., fiscal codes,

inventory codes, interest points addresses) and geo-

coordinates. Since the K-Metropolis DWH is geo-

referenced, the responses to the user queries can be

enriched by graphics and maps.

Any real-time, territorial and semantic DWH

could be structured by either a centralized or

distributed architecture. In the centralized solution,

the semantic data of the DWH would be obtained by

a staging area that transforms the data coming from

the remote SQL tables of the various sources into

triples following ontologies and mappings expressed

by RDF statements (Hayes, 2004).

However, such centralized solution needs a

continuous transformation of the data from the

original archives and their uploading into the urban

DWH to maintain data coherence. Also, this

architecture does not satisfy reliability and privacy

constraints, and places a high computational load on

the central system. For this reason, K-Metropolis

adopted a distributed architecture consisting of a

Cloud of computing nodes, as shown in fig.1, where

the data are kept on the source nodes in both SQL

and RDF formats.

In this architecture the procedures used to

Extract-Transform-Load (ETL) the information

from the distant data sources into the database of

each relevant node can be either off-line or real-time

procedures. The latter approach is mandatory to

update the DBs of the distributed semantic DWH

devoted to control dynamical urban systems, e.g.,

traffic flow optimization by modifying suitably the

traffic lights cycles or first aid interventions by

suggesting the best routes to the ambulances

(Costanzo et al., 2013). In this architecture, the

central server is used only to manage directories,

accounts

and error recovery, whereas the data marts

IntegratingDistributedDataBasesinaSemanticFramework-TheK-MetropolisProject

323

Figure 1: K-Metropolis: functional architecture.

are built by using distributed queries issued in a user

friendly format by a fixed or mobile device. Such

queries are converted into a language, i.e.,

SPARQL 1.1 (Prud’hommeaux and Seaborne,

2008), suitable to extract responses from the triple

stores of the Cloud.

To guarantee a smooth transition from the

current relational archives widely used by the Public

Administration to the envisaged semantic triple

stores, the proposed architecture supports also

queries expressed by distributed SQL formulas so

that it is possible to retrieve data from either SQL

tables or RDF archives.

This choice will allow us to query existing SQL-

based centralized DWH, e.g., as adopted in the

project ELISA promoted by the Italian Ministry for

the Regional Affairs described in http://www.tributi.

eng.it/blog/progetto-elisa/, or to query RDF triple

stores extracted from SQL tables in case the data

owners have defined the controlled vocabulary and

related RDF formats to give semantics to their data.

These ideas have inspired the implementation

architecture of the K-Metropolis project illustrated

in fig.2, where we point out the central XML DB

containing the data taken by a distributed real time

monitoring system dealing with urban transport

infrastructures, e.g., traffic and car pollution, or city

utilities such as electricity, water, gas. Also, XML

and SQL data stores belonging to organizations

offering public services (e.g., parks and pharms) and

Municipal DBs are taken into account to better

support walking and driving citizens.

In particular, fig.2 points out that K-Metropolis

is centered on a Ruby on Rails (RoR) server since

this software environment allows us to implement

the information system per use cases using the

Model-View-Controller paradigm (Hartl, 2011).

The

server is provided with a Decision Support

Figure 2: K-Metropolis implementation architecture.

System (DSS) illustrated in previous works, e.g.,

(Costanzo et al., 2013), to take advantage from all

the urban databases. If the server may enter into the

nodes of the monitoring system, then the DSS may

provide suggestions useful to control specific parts

of the urban infrastructures, e.g. (Aoun, 2013); if the

server may access only the general information

stored on the company servers devoted to manage

the single infrastructure or utility, then the DSS will

provide suggestions to improve daily maintenance,

optimizing the overall costs and planning future

expansions, e.g., (Al-Hader et al., 2009

In both the above cases, the K-Metropolis choice

of using XML protocols to exchange data with the

distant computing systems facilitates the required

DB interoperability, as clarified in sect.2.

The relevant points of interest and possibly the

best paths to reach them are sent to the mobile users

by means of JQMobile scripts (David, 2011) so that

they may be displayed on a Google Maps based user

interface of any mobile.

Moreover, a software based on the Flash Builder

framework (Corlan, 2009) has been developed to

implement the main DSS functionalities on the most

powerful mobiles, e.g., android or iphone devices.

This allows these mobiles to play an autonomous

powerful role in the ubiquitous information system.

Indeed, in this case, the mobiles may carry out e-

government operations without the intervention of

the server by means of their local DSS which takes

into account the personal data stored on the mobiles,

instead of the ones resident on the server, and the

required business data resident on the remote data

bases accessed directly through the GPRS network.

K-METROLIS

INFORMATION SYSTEM

Municipal

DBs

Real Time

Monitoring System

DSS

Flash

Builder

Mobile

DSS

Mobile

XML

RoR Serve

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

324

3 DATA INTEGRATION

As known, the Ontology Web Language (OWL) is a

language for making ontological statements to be

used over the World Wide Web, and all its elements

(classes, properties and individuals) are defined as

RDF resources (Smith et al., 2004).

Unlike traditional systems that operate with a

scoped (or closed) vision of the reality that implies

that a not known statement is automatically false,

OWL makes an open world assumption considering

the unknown information as not explicitly false.

One of the main attractive feature of OWL is that

it allows us to integrate disparate data stores at the

only condition of describing the original data by

OWL statements belonging to a shared

representation of the concepts underlying these data.

Of course, ontologies have an effective utility

depending on if they are agreed by the cooperating

organizations and communities. For this reason, the

activity carried out by the Linked Data group at the

W3C and the Government Linked Data (GLD) that

are publishing data sets and knowledge bases for

supporting e-business and e-government activities

involving different organizations working at global

scale (Bizer et al., 2009) are very important. Such

ontologies may be refined by adding concepts able

to integrate the databases of companies and the ones

of the public departments that operate at local scale.

Although we have not yet well established global

and local ontologies defining e-government and e-

business activities, the mentioned semantic approach

remains the main methodology for the integration of

public data. For this reason, in K-Metropolis we

chosen to integrate the distributed data bases by

transforming the original SQL data into XML

statements to be easily mapped on their turn into the

standard RDF triples, once they will be available,

without changing the codification of the SQL data

stored at the lower layers.

For example, in K-Metropolis the XML

description of the parks of a city that can be

accessed by either the RoR server and the Flash

Builder based applications is as follows:

<parks>

<park>

</park>

</parks>

The only condition to allow the mobiles to retrieve

all the parks is that the RoR server is provided with a

XML directory service that contains the web of the

addresses of the various organizations that offer

parking services, i.e.:

</organization>

</park directory >

Analogous XML descriptions have been defined to

represent the data of the other urban/metropolitan

Public Services. In this way, we may send to the

mobile users all the points of interest (POIs) and

provide the needed services independently on the

organization to which they belong to.

Fig.3 shows the main functionalities offered

currently to the citizens to help their mobility (i.e., to

find nearest convenient “parks” and “fuel stations”),

to assist the search of essential services (i.e.,

“pharms” and in a near future “hospitals”, and “first

aid centres”) and to support typical e-government

operations using the icon “offices”.

Figure 3: Some e-service and e-government functions

offered to the mobile users.

Such functionalities make use of the data

integration technology to offer effective services to

the users, e.g., fig.4 shows how the list of the parks

of different organizations is displayed on the user

IntegratingDistributedDataBasesinaSemanticFramework-TheK-MetropolisProject

325

mobiles after the Flash Builder based application

resident on the mobile has extracted the data

relevant for the user query from different distant

servers using the described XML approach without

any server intervention. Similar information may be

obtained for the fuel stations, the pharms and the

other mentioned essential services.

Figure 4: Parking list meeting the user query. The parks

belong to different organizations.

After having extracted the services required by

the user, the DSS resident on the RoR server or on

the mobile is able to suggest: a) the most suitable

services, b) where they are located on Google Maps,

as illustrated in fig.5, and c) the best path to reach

them, as shown in (Costanzo et al., 2013).

Figure 5: Parking localization displayed on a Google Maps

based representation.

Also, K-Metropolis allows the users to pay in

advance all the mentioned services using PayPal

(fig.6).

Figure 6: e-Payment using PayPal for parking the car in

the park chosen by the user.

Let us note that not only e-commerce activities

but also e-government requests coming from the

citizen mobiles are supported by K-Metropolis, e.g.,

if we press the icon associated to the offices of the

Public Administration in fig.3, the mobile user may

ask certificates (fig.7.left) or may be supported in

compiling an affidavit (fig.7.right). This has been

obtained by transforming the municipal data from

relational tables to XML representations.

Figure 7: e-Certificates, on the left, and e-declaration, on

the right, required from the mobile either directly or

through the K-Metropolis server.

CertificateofResidence

BirthCertificate

Name Surname

Born at Born on

Fiscal_Code

Password

eMail:UserName-Provide

Residence:Cit

-Address

RossiPippo

CataniaIT3‐1‐90

CataniaviaEtnea8,95100

prossigmail.it

(c)

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

326

4 CONCLUSIONS

In the paper we have sketched the project named K-

Metropolis aiming at integrating disparate urban

data bases to support user mobility and to provide e-

commerce and e-government services to desktop

PCs and mobiles.

Although the proposed data integration has been

carried out by filtering a list of XML records

collected from distant sites using simple select-

where like operations, this simple data integration

method is powerful enough to cover many relevant

use cases required by citizens and companies.

Since join operations between distributed data

bases may be useful too, the mentioned companion

paper, i.e., (Giordano et al., 2013), illustrates how K-

Metropolis is able to accomplish this complex task,

as well as it presents how the use of a powerful GIS

may be used to display not only geo-referenced data

of the points of interest but also maps that further

qualify the land use for urban studies or to design

suitable interventions to improve civil protection.

In particular, in this companion paper we show

how using RDF data representations instead of the

XML schemes may improve data integration and

illustrate the SPARQL queries that are able to access

distributed RDF triple stores to carry out both select

and join operations.

Let us note that many approaches have been

proposed to offer information services to mobile

users. The dedicated navigators installed on the cars,

e.g., Garmin and Tom-Tom, were the first examples

of this technology. They may be easily used by the

drivers, but provide only transport information that

don’t take into account very often the real time car

traffic flows.

Although some protocols have been proposed to

improve the real time functionalities of such

navigators, e.g., VANET described in (Offor, 2012),

the car navigators remain with a limited area of

application. In particular, they cannot be used easily

to support walking people mobility, neither can be

used to carry out e-commerce and e-government

tasks.

For this reason, different location based

information systems were proposed in the last years.

They are mainly resident on mobiles and appear to

be a new generation of location based services

(LBSs) that help better people mobility as well as

facilitate e-commerce and e-government operations,

as foreseen in (TRG, 2008).

However, all the proposed LBSs of this second

generation are mainly proprietary systems, so they

miss two basic requirements of the modern LBSs

that are at the basis of the K-Metropolis project, i.e.,

the requirements that the urban data bases should be

open and interoperable, as claimed in (Teller et al.,

2010).

Therefore, our future work will be mainly the

one to study carefully the available urban ontology,

e.g., (Teller et al., 2007; Berdier and Roussey,

2007), to choose the ones that favour the

implementation of an urban presentation layer based

on an standard vocabulary that allows the above

mentioned K-Metropolis applications, resident on

either the server or the mobiles, to access all the

public data available at citywide scale thus

supporting the activity of the mobile users as

completely and flexible as possible.

REFERENCES

Al-Hader, M., et al., SOA of smart city geospatial

management. Third UKSim European Symposium on

Computer Modeling and Simulation, EMS '09, 2009.

Aoun C., The Smart City Cornerstone: Urban Efficiency.

Schneider Electric White Paper, 2013.

Berdier C., Roussey C., Urban Ontologies: the

Towntology Prototype towards Case Studies. Springer,

2007.

Bizer C., Heath T., Berners-Lee T., 2009. Linked data -

the story so far”, Int. J. Semantic Web Inf. Syst.,

vol.5(3).

Corlan M., Flash Platform Tooling: Flash Builder. Adobe,

2009.

Costanzo A., Faro A., Giordano D., Venticinque M., Wi-

City: A federated architecture of metropolitan

databases to support mobile users in real time. Int.

Conf. on Computer and Information Science, ICCIS. A

Conf. of World Engineering Science and Technology

Congress, ESTCON, 2012.

Costanzo A., Faro A., Giordano D., WI-CITY: living,

deciding and planning using mobiles in Intelligent

Cities. 3

International Conference on Pervasive and

Embedded Computing and Communication Systems,

PECCS 2013, Barcelona, INSTICC, 2013.

Crisafi A., Giordano D., Spampinato C., GRIPLAB 1.0:

Grid Image Processing Laboratory for Distributed

Machine Vision Applications. Proc. 17th IEEE Int

Conf on Enabling Technologies: Infrastructure for

Collaborative Enterprises, WETICE ’08, IEEE, 2008.

David M., Developing Websites with jQuery Mobile.

Focal Press, 2011.

Faro A., Giordano D., Musarra A., Ontology Based

Mobility Information Systems. Proc. of Systems, Men

and Cybernetics Conference, SMC’03, vol.3, 4288-

4293, IEEE, 2003.

Faro A., Giordano D., Spampinato C., Evaluation of the

Traffic Parameters in a Metropolitan Area by Fusing

Visual Perceptions and CNN Processing of Webcam

IntegratingDistributedDataBasesinaSemanticFramework-TheK-MetropolisProject

327

Images. IEEE Transactions on Neural Networks, Vol.

19 (6), IEEE, 2008.

Faro A., Giordano D., Spampinato C., Integrating

Location Tracking, Traffic Monitoring and Semantics

in a Layered ITS Architecture. Intelligent Transport

Systems, vol.5(3),IET, 2011.

Faro A., Giordano D., Spampinato C., 2011. Adaptive

background modelling integrated with luminosity

sensors and occlusion processing for reliable vehicle

detection”, IEEE Transactions on Intelligent

Transportation Systems, Vol.12(4).

Giordano D., Torre A., Salemi C., Alessi S., Faro A., An

Ontology based Approach to Integrate Data and Maps

in the Government Enterprise Architecture: a Case

Study. Proc. 15th International Conference on

Enterprise Information Systems, ICEIS, Angers,

INSTICC, 2013.

Hartl M., 2011. Ruby on Rails 3. Addison Wesley.

Hayes. P., 2004. RDF Semantics. W3C Recommendation

10. http://www.w3.org/TR/rdf-mt.

Nebot. V., Berlanga. R., 2012. Building data warehouses

with semantic data. Decision Support Systems,

Vol.52(4). http://sparql-wrapper.sourceforge.net/.

Offor, P.I., Vehicle Ad Hoc Network (VANET). 2012.

Safety Benefits and Security Challenges. Available at

SSRN: http://ssrn.com/abstract=2206077.

Prud’hommeaux E., Seaborne A., 2008. SPARQL Query

Language for RDF. W3C Rec. 15.1.2. http://www.w3.

org/TR/rdf-sparql-query/.

Saha P., 2010. Government Enterprise Architecture

Research Project, NUS Institute of Systems Science.

http:// unpan1.un.org/intradoc/groups/public/

documents/unpan/unpan039390.pdf.

Salampasis, M., Tektonidis, D., Kalogianni, E.,

TraceALL: A Semantic Web Framework for Food

Traceability Systems. Jornal of Systems and

Information Technology, Vol. 14(4), 2005.

Sheth, A., Enterprise applications of semantic web, IFIP

International Conference on Industrial Applications of

Semantic Web. IASW2005, Finland, 2005.

Smith, M. K. , Welty, C.D., McGuinness D.L., OWL Web

Ontology Language Guide, W3C Recommendation.

http:// www. w3.org /TR/ 2004/REC-owl-guide-

20040210, 2004.

Telematics Reserach Group (TRG), Preview portable

navigation: the future is bright for connectivity.

http://www. telematicsresearch.com/PDFs/TRG_Press

_Jan_08.pdf,2008.

Teller J., John R. Lee J.R., Catherine Roussey C., (Eds.),

Ontologies for Urban Development, Springer, 2007.

Teller J., Billen R., Cutting-Decelle A.F., 2010. Ontology

based approaches for improving the interoperability

between 3D urban models. Journal of Inf. Technology

in Construction.

Zhai, J., Jiang, J., Yu, Y., Li J., Ontology-based

Integrated Information Platform for Digital City.

IEEE Proc. of Wireless Communications, Networking

and Mobile Comp., WiCOM ’08, 2008.

ICEIS2013-15thInternationalConferenceonEnterpriseInformationSystems

328