ConvertView

A Tool for Conversion and Visualization of Open Heterogenic Governmental Data

to the RDF Standard

Clóvis H. Nascimento

, Ricardo A. Afonso

, Vinicius C. Garcia

and Carlo M. Revoredo da Silva

Informatics Center – Cin, Federal University of Pernambuco (UFPE), Caixa Postal 7851, 50.740-560, Recife,

Pernambuco, PE, Brazil

Federal University of Alagoas (UFAL), Campus Arapiraca, ARACOMP, Arapiraca, Alagoas, AL, Brazil

Keywords: Open Data, Structured Pattern, Data Extraction, Cloud Computing.

Abstract: The diversity and relevance of the information made available by many governments, using the principles

defined by the open data manifest, creates an opportunity for the populations aid in many fields of

governmental administration, such as security, transportation, health care and education. However, most of

this information was generated in unstructured digital files, making them hard to use. This paper presents an

approach for the structuring of data, improving the processing and extraction of information. Thus, we created

a tool based on a cloud computing environment, which has high processing and storage ability, to create files

in a homogeneous and structured format, allowing for ominous and specific queries; contributing to the access

of the information.

1 INTRODUCTION

When the manifesto named Open Data (Accar 2009)

appeared through the governments of many countries

like United States and Brazil, to provide their legacy

and current data to the world through the internet,

created positive expectations in the population. The

manifesto aimed, total and permanent transparency of

their data produced in several areas, such as security,

transportation, health, education, among others. In

addition the use of these data, enables a better

political and economic contribution from the

population, for example, identifying the areas that

require a greater investment of resources by the

government.

In Brazil’s case, these expectations were not fully

met, not because of the data itself, but because of the

difficulty in using them. According to (Breitman, K.

et al. 2011) our biggest challenge today comes from

the diversity of data sources and the provenance of

these sources. Furthermore, most of the data provided

by the Brazilian government are not in the Linked

Data standard (Bizer, C. 2009), which allows the

interconnection of different data sources, being the

format recommended by the W3C consortium. The

available Open data is unstructured (documents,

spreadsheets etc.), despite several government

initiatives supported by the W3C, the results aren’t

sufficient to define a consistent approach, which

would facilitate the creation of an infrastructure of

Governmental Data in the Linked Data standard.

Despite the listed problems, the exponential

growth in available data on the web and other open

data initiatives creates new challenges, such as the

processing of large volumes of information and

minimizing the high costs of hardware and software.

To solve these problems, new technological

approaches emerged, such as big Data (Wu et al.

2013), designed to process large volumes of

distributed, heterogeneous data and cloud computing

(Qi Zhang et al. 2010), which was used in this work.

Cloud Computing provides hardware and software as

a service for a low cost combined with various

benefits such as high processing power, storage,

scalability and so on. According to (Simmhan et al.

2010), scalability and ease of integrating data from

multiple heterogeneous networks, justifies the

adoption of cloud computing.

In this context, this paper aims to contribute in the

research of heterogeneous processing for the RDF

(Resource Description Framework) data standard,

thus, in an attempt to fill this gap, we propose a tool

that operates in the cloud computing environment

called ConvertView containing options for the user to

253

H. Nascimento C., A. Afonso R., C. Garcia V. and Revoredo da Silva C..

ConvertView - A Tool for Conversion and Visualization of Open Heterogenic Governmental Data to the RDF Standard.

DOI: 10.5220/0005456902530258

In Proceedings of the 11th International Conference on Web Information Systems and Technologies (WEBIST-2015), pages 253-258

ISBN: 978-989-758-106-9

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

convert XML file to RDF, SPARQL queries and

display (Simple Protocol and RDF Query Language),

in the destinations files. It is expected to evolve to the

state of art in the literature.

The paper was divided into six sections, starting

with the introduction to the subject, describing the

problem and the motivation. Section 2 presents the

related work, section 3 presents the proposed solution

and its implementation, section 4 presents the results,

section 5 the conclusion and finally in section 6 the

references.

2 TOOLS AND RELATED WORK

This section describes briefly some recent references

on the state of the art, since many works have the

same goal: to solve the problems of data

transformation for semantic format, but follow

different approaches.

A tool called StdTrip, developed by (Salas, P. et

al. 2011), generates RDF files from relational

databases of the Brazilian government. This approach

requires the triplication of the data set, in other words,

the conversion to RDF of database schemas and their

instances. A key issue in this process is deciding how

to represent concepts of database schema in terms of

RDF classes and properties by mapping database

concepts and an RDF vocabulary.

The Rich Internet Application (RIA) called

WebSmatch, created by (Coleta, Remi. et al. 2012),

supports the entire process of importing, refining and

integrating the data sources and uses third-party tools

for high quality display. This application uses a

common situation for the integration of public data

unresolved by current tools: the input files are poorly

structured (excel xls in this case) and occurs only a

superficial view of these data.

(Edgard Marx et al. 2012) Created a tool that

generates RDF files, database schemas and mappings,

providing a plug-in for use with the Eclipse program

and connects to the DBMS via JDBC (Java Database

Connectivity). The tool offers a range of mapping

languages, including R2RML (RDB to RDF Mapping

Language), however it does not support SPARQL

queries to perform.

The tool implemented by (Giacomo, G. et al.

2012) provides access to any data management

system through the use of ontologies for semantically

representing the data source. The connections with

the DBMS are made via JDBC and semantic

mappings are created using OWL (Web Ontology

Language). The tool called Mastro provides a tool

called OBDA environment (Ontology Based Data

Access), with resources for creating and querying, as

well as enabling consistency and reasoning checking

on OWL ontology.

Despite the quality and relevance of the work

described in this section, they do not use cloud

computing. It´s important to recall that according to

(Mutavdzic 2010) cloud computing can reduce costs,

simplify management, improve service, provide

transparency and allow interaction with the flexible

citizen, helping to fulfil their governments wish to

transform the way they serve their citizens.

Government agencies and departments have

difficulty to identify and implement opportunities for

shared services that have a good chance of

successfully implementing new services to citizens.

One of the advantages of the proposed tool is the

ability to query the RDF files without the necessity of

previous knowledge of the SPARQL query language.

The intuitive interface allows anyone to use filters

that will generate the query, and the tool shows the

query command created by the filter selection. Table

1 shows a comparison of the tools described in this

paper:

Table 1: Related works comparative table.

Project Objective Entry

StdTrip

Generate RDF,

database schemes and

mappings. Suggests

vocabulary for the

concepts.

Database schemes

WebSmatch

Generate XML and

CVS files from XLS

Excel sources.

XLS Excel

RDB2RDF

Create RDF,

database schemes

and mappings.

Provides plugins to

use with Eclipse and

connects the bases

through JDBC

Connection with the

database through

JDBC

Mastro

Generate a well-

structured file from a

database

Relational database

ConvertView

Generate RDF,

mappings,

visualization and

consultation from

sources

WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies

254

Table 1: Related works comparative table (cont.).

Automation

Level

Mapping

Language

Supports

SPARQL

Using

Cloud

Computing

Manual/

Automatic

SQL,

RDF,

R2RML

No No

Manual XML No No

Manual/

Automatic

R2RML No No

Automatic OWL No No

Manual XML Yes Yes

3 PROPOSED SOLUTION

In this section, a proposal for solving the problems

described above is presented. For this reason, a tool

that enables users to generate RDF databases from

datasets in the XML (eXtensible Markup Language)

format was created. The process of creating the RDF

file is shown in Figure 1, starting with the user

selecting a dataset file that served as basis for RDF.

Then the terms in the datasets are presented on the

screen and the user chooses the words and the

corresponding vocabularies, where the available

vocabularies are FOAF (Friend of a Friend) and DC

(Dublin Core), and finally the RDF file and mapping

are generated.

Figure 1: Steps mapping and generation of RDF file.

In state of art research, the majority of studies found

use as a base XML, XLS, databases and other formats

for the generation of RDF files. We observed that all

of them operate interactively in the selection of the

used vocabularies, making it up to the user to decide

the vocabulary used in the process of generating RDF

files. Another observation made was related to the

languages of mappings: there is no standard, ie, each

job uses a different representation formats that are

worth highlighting: XLS, SQL, RDF and OWL. The

consortium W3C (World Wide Web Consortium)

recommends the use of the mapping language

R2RML (RDB to RDF Mapping Language) since late

2012.

3.1 Implementation

The development of a tool for generation of semantic

data, depending on the complexity of data sources and

the degree of formality to be generated, can be an

arduous and lengthy process. However, the previous

clear definition of the steps should to be followed

until their conclusion, in order to facilitate their

implementation.

After defining the purpose and scope of the tool,

the technologies needed for implementation are

chosen. In this section, the technologies and

development of the environment, which were used in

the creation of the ConvertView tool, was presented.

3.2 Used Technology

The programming language and computing platform

Java, version 1.7, was chosen because of its

scalability and multi-platform aspects. In this work

Jena Semantic Web Framework 2.7.4 was used to

generate RDF files and perform queries SPARQL

(Protocol and RDF Query Language) endpoint.

3.3 Development Environment

The chosen Open Source environment was the IDE

(Integrated Development Environment) provided by

Eclipse, used in this work to write, debug, and run the

source code of the ConvertView tool. The Google App

Engine platform offered as a service (PaaS) of cloud

computing for the development of web applications,

enabled the creation, implementation and management of

the developed application developed and was chosen to

provide plugins for the Eclipse IDE.

3.4 System Architecture

Observing the architecture of the tool in Figure 2, it is

possible to observe that it has only two layers, one for

the client and one for the cloud server. In the diagram

we can see the main components of each layer and its

interfaces.

4 RESULTS

In this section, the results obtained from the use of the

Convertview tool will be presented. For this, some

datasets that were used and various combinations of

terms and vocabularies were selected, in order to

demonstrate the use and efficacy of the tool. As

shown in Figure 3, a dataset called Professores.xml

was selected and the tool showed the terms presented

ConvertView-AToolforConversionandVisualizationofOpenHeterogenicGovernmentalDatatotheRDFStandard

255

Figure 2: Architecture Diagram of ConvertView Tool.

in the dataset (in this case, Email, Matricula, Nome

and Sala). Then, the user picked the three first terms,

informing the desired vocabularies (accountname,

identifier and name).

Figure 3: Main Screen of the tool ConvertView.

After selecting the terms and vocabularies, clicked in

the button generate, to make possible complete the

process of creating the RDF. Figures 4 and 5 show the

mapping files and RDF file, respectively.

<?XML version="1.0" encoding="UTF-8"

standalone="no"?>

<mappings> <mapping> <term>Email</term>

<vocabulary>accountname</vocabulary>

<term>Matricula</term>

<vocabulary>identifier</vocabulary>

</mapping></mappings>

Figure 4: Mapping File generated by the ConvertView tool.

<RDF:RDF

XMLns:RDF="http://www.w3.org/1999/02/22-RDF-

syntax-ns#"

XMLns:j.0="http://XMLns.com/foaf/0.1/"

XMLns:dc="http://purl.org/dc/elements/1.1/">

<RDF:DescriptionRDF:about="http://www.cin.ufpe.br

/thing/Professor2">

<j.0:name>Vinicius Cardoso Garcia</j.0:name>

<dc:identifier>2</dc:identifier>

<j.0:accountName>vcg@cin.ufpe.br</j.0:accountNam

</RDF:Description>

<RDF:DescriptionRDF:about="http://www.cin.ufpe.br

/thing/Professor1">

<j.0:name>Silvio Romero de LemosMeira</j.0:name>

<dc:identifier>1</dc:identifier>

<j.0:accountName>srlm@cin.ufpe.br</j.0:accountNa

me>

</RDF:Description>

</RDF:RDF>

Figure 5: RDF file generated by the ConvertView tool.

The mapping file was created in the XML format,

since semantic expressiveness with logical

constructors is not needed as it is in the OWL

language, with simple correspondence of terms and

vocabularies taking its place.

After creating the RDF file, it is also possible to

check your data inside the tool, using the SPARQL

language, as shown in Figure 6.

The tool itself generates the prefixes of the

vocabularies and the user can enter any“select”

commandto search in the generated RDF file. Figure

7 shows the use of the select command, which showes

on the screen all the data contained in the RDF file

sorted by name.

WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies

256

5 CONCLUSIONS

This work aimed to advance the state of the art

concerning the limitations of use of open data, due to

the absence of structured modelling. For this, the

Convertview tool was developed, showing several

advantages due to its approach in a cloud computing

environment, with high processing power,

availability, storage and scalability. A summary of the

issues involved in its development and use, the

technologies and platforms used, as well as the

reasons these choices was exposed.

The tests made allowed to confirm the success in

the generation of RDF files and the option to make

SPARQL queries complements the usefulness of the

tool, which eliminates the need for another program

to search the generated data. However, the tool is still

in its early stage of development and the next steps of

development and maturity predict the following

developments in the following software:

1. Formats dataset sources: Increase the amount of

file types it is able to read (cvs, json, xls, etc), not just

XML;

2. Increase the amount of vocabularies: Use also other

famous and stable vocabularies, such as EVENT,

Geonames, AKT, BIBO, SWC and SWRC with all its

terms, to include a wider variety of areas;

3. Suggestions of terms and vocabularies: Suggest

terms and vocabularies, based on the content of the

dataset, assisting in the choice of the user;

4. Mapping Languages: Add the option to choose the

mapping language, to allow better interoperability

with other tools.

5. Allow the creation of RDFS: If the user wishes,

RDFS files could be generated as well, with classes

and subclasses defined by the user.

6. Allow the input of simultaneous sources: Enable

the reading of various heterogeneous sources

simultaneously, allowing to correlate equivalent

terms in order to enrich future consultations.

Finally, it is evident that the demand for such tools

has increased due the amount of unstructured data

that has been provided, especially on internet

files,thus revealing that there is much work to be

done. In this context, this work was developed to help

to fill this gap.

Figure 6: Screen filter search automatic or manual.

ConvertView-AToolforConversionandVisualizationofOpenHeterogenicGovernmentalDatatotheRDFStandard

257

Figure 7: Screen queries in SPARQL.

REFERENCES

Accar, S., Alonso, J., Novak, K. (eds). Improving Access to

Government through Better Use of the Web (2009).

W3C Interest Group.

Bizer, C. “The Emerging Web of Linked Data”; IEEE

Intelligent Systems (Sep/Oct. 2009).

Breitman, Karin K., Salas P., Saraiva D., Gama V.,

Casanova Marco A., Open Government Data in Brazil,

Department of Informatics, Pontifícia Universidade

Católica do Rio de Janeiro, Rio de Janeiro, 2011.

Coleta, Remi.,Castanier, E., Valduriez P., Frisch C., Ngo,

D., Belllahsene Z., Public Data

IntegrationwithWebSmatch, INRIA and LIRMM, Data

Publica, Frannce, 2012.

Edgard Marx, Percy Salas, Karin Breitman, José Viterbo

and Marco Antonio Casanova, RDB2RDF: A

relationalto RDF plug-in for Eclipse, Informatics

Department, Pontifícia Universidade Católica do Rio de

Janeiro, 2012.

Giacomo, G.,Lembo, D., Lenzerini, M., Poggi, A. Riccardo

Rosati, Ruzzi, M., Savo, D. F, Mastro: A Reasoner for

EfectiveOntology-Based Data Access, University

Sapienza of Roma, Italia, 2012.

MUTAVDŽIC, R. “Cloud Computing Architectures for

National, Regional and Local Government”.Microsoft

Croatia. 2010.

Qi Zhang Lu Cheng Raouf Boutaba The Brazilian

Computer Society. Cloud computing: state-of-the-art

and research challenges, 2010.

Salas, P., Viterbo, J., Breitman, K. and Casanova, M.A.

StdTrip:Promoting the Reuse of Standard Vocabularies

in Open Government Data. In: D. Wood (ed.) Linking

Government Data, Springer Verlag (2011).

Simmhan, Y., Giakkoupis, M., Cao, B., Prasanna, V. K.

(2010) “On Using Cloud Platforms in a Software

Architecture for Smart Energy Grids,” IEEE

International Conference on Cloud Computing.

Wu, X., Zhu, X., Wu, G.-Q., and Ding, W. (2013). Data

mining with big data. IEEE Transactions on Knowledge

and Data Engineering, 99(PrePrints):1.

WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies

258