Web portals are currently an important means to access Internet information. The use of Web portals permits a vast amount of data to be obtained rapidly. However, the quality of the data recovered by the user is fundamental. We therefore propose a thesis in which a model, denominated as SPDQM (SQuaRE Portal Data Quality Model) will be defined. The proposed model will be based on a previous model, PDQM (Portal Data Quality Model) and the SQuaRE (Software product Quality Requirements and Evaluation) standard. Finally, upon the model's completion, an automatic tool with which to asses the data quality (DQ) in Web portals will be developed.
A Web portal is a Website or service that offers a broad array of resources and
services for customers and business partners [
In order to develop this model it is first necessary to study the existing research into data quality in general and the data for Web portals in particular.
With regard to data quality, SQuaRE came into existence some months ago. SQuaRE was selected in this paper, because it is the most recent series of International Standards in which a data quality model is defined. We therefore believe that it is necessary to consider the quality characteristics identified in this standard.
With regard to data quality in Web portals, we should highlight the PDQM model. However, its definition is previous to SQuaRE, it is specific to only one type of Web portal: that of universities, and it has only been partially implemented.
PDQM and SQuaRE will therefore be used as a basis to develop a more complete DQ model for Web portals.
This article is organized as follow. In Section 2, a background is presented, while the thesis proposal is presented in Section 3. 2
2.1 PDQM (Portal Data Quality Model)
As was previously mentioned, the PDQM (Portal Data Quality Model) model will be
used as a starting point. PDQM is focused on the perspective of the data consumer.
The development of PDQM was divided into two stages: the theoretical definition and
the operational definition of the model [
The goal of the theoretical definition was to determine a set of DQ characteristics that are relevant to data consumers when evaluating the DQ of any Web portal. To do this, a set of DQ characteristics proposed in literature was chosen to evaluate DQ in a Web context and a selection of the most relevant characteristics for a Web portal was then defined. This set was empirically validated, resulting in the final set of DQ characteristics for the model. The operational version of PDQM was obtained by first organizing the characteristics into four DQ categories: - Intrinsic, which denotes that data have quality in their own right. - Operational, which emphasizes the importance of the role of systems; that is, the system must be accessible but secure. - Contextual, which highlights the requirement which states that data quality must be considered within the context of the task in hand. - Representational, which denotes that the system must present data in such a way that they are interpretable, easy to understand, and concisely and consistently represented.
In each category, influential relationships were then established between the
characteristics to determine which characteristics were dependent on other characteristics.
As a result of this, a BN (Bayesian Network) was obtained which organizes the 33
DQ characteristics into four network fragments (one for each DQ category). The
problem is that PDQM was only created for university Web portals and, up to now,
only evaluates the representational category.
SQuaRE is a set of International Standards which consists of different divisions. Our
model will be based on the ISO/IEC 25012 which proposes a data quality model that
defines fifteen characteristics considered from two points of view: inherent and
system dependent [
Inherent data quality refers to the degree to which quality characteristics of data
have the intrinsic potential to satisfy stated and implied needs when data is used under
specified conditions [
- Problems to solve: As was explained in the introduction, Internet has increased the
amount of data that can be obtained through the network and the ability to provide
and obtain information from several sources. Web portals thus serve as an
important means to access information. Web portals have therefore undergone an
evolution and currently provide a variety of services. One of the aims of many
Web portals is to select, organize and distribute content (information or other
services and products) in order to satisfy their users/customers [
Survey: The set of data quality characteristics which are relevant to the existing
proposal have been obtained through a survey, based on the steps of
Kitchenham [
Methodology to develop a quality model: the quality model will be defined by using a methodology which is being developed by colleagues at the University of Castilla-La Mancha and the University of Málaga.
Methodology to define the measures: [
This methodology will be used to define the necessary measures. - Model: The model was developed by using both the set of characteristics proposed in PDQM and ISO/IEC 25012 and those obtained as a result of our survey. Due to the survey selects the DQ characteristics for Web in general, the next step is to refine the set of obtained characteristics and to study the applicability of the characteristics to the Web portal context. Once the characteristics which are most suitable for Web portals had been chosen it was necessary to resolve conflicts. This was done by detecting both those characteristics which have the same name but a different meaning and those characteristics which have a different name but refer to the same meaning. And when a characteristic has an only subcharacteristic, the subcharacteristic is removed and it is taken into account in the definition of the characteristic. As a final result, we have obtained 42 DQ characteristics (see Table 1).
Having identified all the characteristics there are two possibilities to continue with the operationalization of SPDQM:
We can classify the characteristics according to ISO/IEC 25012 as being “Inherent” and “System Dependent”. We would therefore concentrate on both (“Inherent” and “System Dependent”), and would define measures for each of its characteristics. For some characteristics, the measures would be derived from the user’s opinions. However, we would like that the vast majority of the measures were automatable to allow us to develop a tool which is capable of evaluating a Web portal. The tool will provide us with values for each indicator that will have been defined (one for each characteristic). The tool will also permit the user to give more or less importance to each indicator. For example, for each indicator it is possible to ask the user if s/he believes it to be very important, not very important or not at all important. For the subjective measures, the tool will ask to the users, their opinion. The tool will use these values to calculate the final value of the characteristics associated with the “Inherent” and the “System Dependent” points of view.
The second option is to attempt to create an extension of PDQM that is aligned with SQuaRE. To do this we would consider the “inherent” point of view of the ISO/IEC 25012 which corresponds with PDQM’s intrinsic category, and “System Dependent” which corresponds with the other three categories (see Table 1). Therefore, by following the creation of a bayesian network as that of PDQM, we would have to add new characteristics and group them with a maximum of three entries per node. The network would thus have a first and second level with the identified characteristics and the necessary artificial nodes. A third level would contain the PDQM categories, and a fourth level would contain those of ISO/IEC 25012.
We have not yet decided which option is most suitable.
Automatic tool: We will create an automatic tool to evaluate the DQ in Web portals. This tool will calculate the level of DQ based in measures calculated automatically from the web portal and other measures obtained from user´s input (like a questionnaire). The goal of this automatic tool is to facilitate the developer’s task when incorporating data quality into Web portals and to provide the users with the DQ level of a specific Web portal.
Contributions to Web Engineering: This model could be used by Web portal designers to improve data quality, promote good practices in DQ on the Web, come up with data cleaning techniques, develop patterns for data refinement or eliminate unnecessary data. Thus, the automatic tool will be developed to provide us with the necessary guidelines to carry out those improvements.
Acknowledgments. This work is part of the INCOME project (PET2006-0682-01) supported by the Spanish Ministerio de Educación y Ciencia, by the IVISCUS project (PAC08-0024-5991) supported by Consejería de Educación y Ciencia (JCCM) and by VIMECUS (TC20080556) supported by the University of Castilla-La Mancha. PhD Student: Carmen Moraga; Supervisors: Mª Ángeles Moraga, Angélica Caro